Unit – I

Meaning, Scope, Functions and Limitations of Statistics

Statistics Definition, Importance, Scope, Limitation

Statistics is a branch of mathematics that deals with collecting, analyzing, interpreting, presenting, and
organizing data. It provides methods to describe and quantify variability in data in a way that allows for informed
decision-making. Statistics can be descriptive, summarizing data points using metrics like mean and standard
deviation, or inferential, making predictions and testing hypotheses based on sample data extrapolated to larger
populations. It is applied across various fields such as economics, psychology, medicine, business, and
engineering to help stakeholders understand trends, make forecasts, optimize processes, and evaluate
theories. By converting raw data into meaningful information, statistics enables deeper insights and supports
the decision-making process under conditions of uncertainty.

Importance of Statistics:

• Informed Decision Making: Statistics provides a basis for making decisions based on data rather than
assumptions or intuition. This is particularly crucial in business, economics, and healthcare where
strategic decisions need to be data-driven.

• Understanding of Trends: Statistics helps identify patterns and trends within large sets of data. This is
essential for market analysis, economic forecasting, and social sciences, helping to predict future
occurrences based on historical data.

• Quality Testing: In manufacturing and production, statistics are used to maintain quality control.
Statistical methods are applied to ensure that products meet quality standards consistently, minimizing
the risk of defects.

• Risk Assessment: Statistics allow businesses and investors to assess the potential risks involved in
various decisions, providing a quantitative basis for risk management strategies in finance, insurance,
and more.

• Efficiency Improvement: By analyzing data statistically, organizations can identify areas of process
improvement, optimize operations, and increase overall efficiency. This is evident in sectors like
logistics, manufacturing, and service delivery.

• Policy Development: In public policy and government, statistics are essential for the development and
evaluation of policies. Statistical data supports policy decisions and helps measure the impact of policy
interventions.

• Academic Research: In academia, statistics is vital for validating research findings and hypotheses. It
supports the credibility of research through precise measurement, analysis, and interpretation of data.

• Healthcare Advances: Statistics are critical in healthcare for determining the effectiveness of
treatments, understanding patient outcomes, and conducting epidemiological studies, which guide
public health decisions and medical practices.

Scope of Statistics:

• Business and Economics: Statistics are used to analyze market trends, consumer behavior, and
economic conditions. It aids in risk management, financial forecasting, marketing strategies, and
decision-making processes.

• Healthcare: In the medical field, statistics is critical for understanding the effectiveness of treatments,
managing patient data, designing clinical trials, and making policy decisions on public health.

1|Page
 • Government and Public Policy: Statistical data is fundamental in formulating public policies,
conducting census activities, planning urban development, and evaluating the impact of policy
decisions.

• Environmental Science: Statistics help in environmental monitoring, assessing pollution levels,

predicting changes in climate patterns, and studying biodiversity. It’s crucial for developing sustainable
practices.

• Education: In education, statistics are used to analyze student performance, evaluate teaching
methods, and research educational theories. It also plays a role in institutional planning and
management.

• Social Sciences: For fields such as psychology, sociology, and anthropology, statistics is indispensable
for conducting research, testing theories, and understanding complex social phenomena.

• Engineering: Statistical methods are utilized in engineering for quality control, product design, reliability
testing, and optimization of processes.

• Sports: Statistics are increasingly used in sports for player analysis, game strategies, and team
performance evaluations, enhancing both the spectator experience and team management.

• Information Technology and Data Science: In the age of big data, statistics is fundamental in data
mining, machine learning algorithms, and the processing and interpretation of massive data sets for
actionable insights.

• Finance and Insurance: Statistics underpin financial analysis, portfolio management, insurance risk
assessment, and predictive modeling to forecast market movements and mitigate risks.

Limitation of Statistics:

• Misinterpretation Risks: Statistics can be misinterpreted or misused, intentionally or unintentionally.

Simplified conclusions drawn from complex data can lead to misleading representations of reality.

• Dependence on Quality of Data: The accuracy of statistical analysis is heavily dependent on the quality
of the data used. Poor data, whether due to error in data collection or biased samples, can lead to
incorrect conclusions.

• Lack of Specificity: Statistical analysis often deals with generalizations and may not apply to individual
cases or specific instances. This can be particularly limiting in fields like medicine or personalized
services where individual variation is significant.

• Complexity in Data Collection: Gathering comprehensive and relevant data for statistical analysis can
be expensive, time-consuming, and technically challenging, which can limit the scope and reliability of
the studies.

• Probability and Uncertainty: Many statistical methods are based on probability, which introduces an
element of uncertainty in predictions and forecasts. This uncertainty can sometimes be significant,
depending on the variability of the data.

• Inapplicability to Qualitative Data: Traditional statistical methods are geared towards quantitative data
and often cannot be used to analyze qualitative information, such as opinions or feelings, which can be
equally important in some research contexts.

• Influence of Outliers: Extreme values or outliers can disproportionately influence the results of
statistical analysis, leading to skewed data interpretations unless specifically accounted for in the
analysis process.

2|Page
 • Requires Statistical Expertise: Effective use of statistics requires a certain level of expertise.
Misapplication of statistical methods or inappropriate choice of techniques can lead to errors in
conclusions, making it necessary for practitioners to have a solid understanding of statistical principles.

Measures of Central Tendency: Mean, Median, Mode

Measures of Central Tendency: Mean, Median, and Mode

Measure of Central tendency is a summary statistic that represents the center point or typical value of a
dataset. These measures indicate where most values in a distribution fall and are also referred to as the central
location of a distribution. You can think of it as the tendency of data to cluster around a middle value. In
statistics, the three most common measures of central tendency are the mean, median, and mode. Each of
these measures calculates the location of the central point using a different method.

The mean, median and mode are all valid measures of central tendency, but under different conditions, some
measures of central tendency become more appropriate to use than others. In the following sections, we will
look at the mean, mode and median, and learn how to calculate them and under what conditions they are most
appropriate to be used.

Mean (Arithmetic)

The mean (or average) is the most popular and well known measure of central tendency. It can be used with both
discrete and continuous data, although its use is most often with continuous data (see our Types of
Variable guide for data types). The mean is equal to the sum of all the values in the data set divided by the number
of values in the data set. So, if we have n values in a data set and they have values x 1, x2, …, xn, the sample mean,
usually denoted by (pronounced x bar), is:

This formula is usually written in a slightly different manner using the Greek capitol letter, , pronounced “sigma”,
which means “sum of…”:

You may have noticed that the above formula refers to the sample mean. So, why have we called it a sample
mean? This is because, in statistics, samples and populations have very different meanings and these
differences are very important, even if, in the case of the mean, they are calculated in the same way. To
acknowledge that we are calculating the population mean and not the sample mean, we use the Greek lower
case letter “mu”, denoted as µ:

The mean is essentially a model of your data set. It is the value that is most common. You will notice, however,
that the mean is not often one of the actual values that you have observed in your data set. However, one of its
important properties is that it minimizes error in the prediction of any one value in your data set. That is, it is the
value that produces the lowest amount of error from all other values in the data set.

An important property of the mean is that it includes every value in your data set as part of the calculation. In
addition, the mean is the only measure of central tendency where the sum of the deviations of each value from
the mean is always zero.

3|Page
Median

Median is the middle score for a set of data that has been arranged in order of magnitude. The median is less
affected by outliers and skewed data. In order to calculate the median, suppose we have the data below:

65 55 89 56 35 14 56 55 87 45 92

We first need to rearrange that data into order of magnitude (smallest first):

14 35 45 55 55 56 56 65 87 89 92

Our median mark is the middle mark – in this case, 56 (highlighted in bold). It is the middle mark because there
are 5 scores before it and 5 scores after it. This works fine when you have an odd number of scores, but what
happens when you have an even number of scores? What if you had only 10 scores? Well, you simply have to
take the middle two scores and average the result. So, if we look at the example below:

65 55 89 56 35 14 56 55 87 45

We again rearrange that data into order of magnitude (smallest first):

14 35 45 55 55 56 56 65 87 89

Only now we have to take the 5th and 6th score in our data set and average them to get a median of 55.5.

Mode

The mode is the most frequent score in our data set. On a histogram it represents the highest bar in a bar chart
or histogram. You can, therefore, sometimes consider the mode as being the most popular option. An example
of a mode is presented below:

4|Page
 Quartiles
Partition Values: Quartile, Deciles, Percentiles

Partition Values are statistical measures that divide a dataset into specific parts, helping in understanding the
distribution of data. These include quartiles, deciles, and percentiles.

Quartiles: Quartiles divide the dataset into four equal parts, each containing 25% of the data points.

Key Quartiles:

• Q1 (First Quartile): The value below which 25% of the data lies.

• Q2 (Second Quartile/Median): The value below which 50% of the data lies.

• Q3 (Third Quartile): The value below which 75% of the data lies.

Formula: Qk = k(n+1) / 4

Where

k is the quartile position (1, 2, or 3)

n is the number of observations.

Deciles: Deciles divide the dataset into ten equal parts, each containing 10% of the data points.

Key Deciles:

• D1 (First Decile): The value below which 10% of the data lies.

• D5 (Fifth Decile): The value below which 50% of the data lies (this is also the median).

• D9 (Ninth Decile): The value below which 90% of the data lies.

Formula: Dk = k(n+1)10

Where k is the decile position (1 to 9).

Percentiles: Percentiles divide the dataset into 100 equal parts, each containing 1% of the data points.

Key Percentiles:

• P1 (First Percentile): The value below which 1% of the data lies.

• P50 (Fiftieth Percentile): The value below which 50% of the data lies (this is also the median).

• P99 (Ninety-ninth Percentile): The value below which 99% of the data lies.

Formula: Pk = k(n+1) / 100

Where k is the percentile position (1 to 99).

Measures of Dispersion: Range, Inter Quartile Range

Measures of Variation: Range, IQR

In statistics, measures of variation, also known as measures of dispersion, provide insights into how data points
spread or deviate from the central tendency (mean, median, or mode). Two commonly used measures of
variation are Range and Interquartile Range (IQR). Understanding these concepts is crucial to gaining a deeper
comprehension of the dataset’s variability, which is critical in decision-making and analysis.
5|Page
Range: The range is the simplest measure of variation and is calculated as the difference between the largest
and smallest values in a dataset.

Range = Max value − Min value

• Example: Consider a dataset representing the ages of individuals in a group: 15, 18, 21, 24, 30, and 35.
The range would be:

Range = 35−15=20

Thus, the ages span 20 years.

• Interpretation: The range provides an understanding of the extent of the dataset. It gives a basic sense
of how spread out the data points are. A larger range indicates a greater spread, while a smaller range
suggests that the data points are closer together.

• Limitations: The range only considers the two extreme values (minimum and maximum) and ignores the
distribution of the other data points. It is sensitive to outliers. For example, if the highest or lowest value
is an extreme outlier, the range could be misleading.

• Use Cases: Range is useful in quick, rough assessments of variability. It is commonly used in simple
comparisons between datasets, such as measuring temperature fluctuations over a day, price ranges of
products, or age differences in a group.

Interquartile Range (IQR):

The Interquartile Range (IQR) is a more robust measure of variation, calculated as the difference between the
third quartile (Q3) and the first quartile (Q1) of the dataset. The IQR measures the spread of the middle 50% of
the data, which excludes outliers and extreme values.

IQR = Q3−Q1

Quartiles:

• Q1 (First Quartile): The value below which 25% of the data lies.

• Q3 (Third Quartile): The value below which 75% of the data lies.

• Together, Q1 and Q3 provide the boundaries for the middle 50% of the dataset.

Example: Consider a dataset of exam scores: 55, 60, 65, 70, 75, 80, 85, 90, and 95. The quartiles would be:

• Q1 = 65 (first quartile)

• Q3 = 85 (third quartile)

The IQR would be: IQR = 85 − 65 = 20

This means that the middle 50% of the exam scores range between 65 and 85.

• Interpretation: IQR is a more accurate measure of variability when compared to the range, especially in
skewed distributions or datasets with outliers. Since it focuses on the middle 50% of data, it avoids being
distorted by extremely high or low values. A larger IQR indicates a greater spread in the middle portion of
the data, while a smaller IQR suggests that the central data points are closely grouped.

• Box Plot and IQR: IQR is visually represented in box plots, where the box itself reflects the range between
Q1 and Q3. The “whiskers” of the box plot extend to the minimum and maximum values, while the box
highlights the central 50% of the data.

6|Page
 • Outliers and IQR: Outliers are typically identified using the IQR. Any value that lies more than 1.5 times
the IQR above Q3 or below Q1 is considered an outlier. This method helps in distinguishing between
normal data variation and unusual data points.

Lower Bound=Q1−1.5×IQR

Upper Bound=Q3+1.5×IQR

Any values outside these bounds are flagged as potential outliers.

• Use Cases: The IQR is especially valuable in descriptive statistics, particularly when dealing with
skewed data or data with outliers. For example, it is frequently used in financial analysis to measure
variability in income or asset distributions, in education to analyze test scores, or in research to evaluate
the spread of experimental data.

Range vs. IQR: Key Differences

• Range provides a quick sense of the total spread of the data, but it is highly sensitive to extreme values.

• IQR, on the other hand, focuses on the central portion of the data and is much more robust against
outliers.

While the range is easier to compute and offers a rough estimate of variability, the IQR provides a more nuanced
understanding of the dataset’s dispersion, especially when there is concern about skewed data or outliers.

Mean Deviation, Standard Deviation

Mean Deviation and Standard Deviation

Mean Deviation

To understand the dispersion of data from a measure of central tendency, we can use mean deviation. It comes
as an improvement over the range. It basically measures the deviations from a value. This value is generally
mean or median. Hence although mean deviation about mode can be calculated, mean deviation about mean
and median are frequently used.

Note that the deviation of an observation from a value a is d= x-a. To find out mean deviation we need to take the
mean of these deviations. However, when this value of a is taken as mean, the deviations are both negative and
positive since it is the central value.

This further means that when we sum up these deviations to find out their average, the sum essentially vanishes.
Thus, to resolve this problem we use absolute values or the magnitude of deviation. The basic formula for finding
out mean deviation is:

Mean deviation= Sum of absolute values of deviations from ‘a’ ÷ The number of observations

Standard Deviation

As the name suggests, this quantity is a standard measure of the deviation of the entire data in any distribution.
Usually represented by s or σ. It uses the arithmetic mean of the distribution as the reference point and
normalizes the deviation of all the data values from this mean.

Therefore, we define the formula for the standard deviation of the distribution of a variable X with n data points
as:

7|Page
 Variance, Coefficient of Variance
Variance, Coefficient of Variance

Variation refers to the degree of change or difference in data points within a dataset. It measures how values
differ from each other or the central tendency (mean or median). Variation helps understand data spread,
identifying whether values are tightly clustered or widely dispersed. Key measures of variation include range
(difference between maximum and minimum values), variance (average squared deviation from the mean), and
standard deviation (square root of variance). High variation indicates diverse data, while low variation suggests
uniformity, aiding in data analysis and decision-making.

formula for the coefficient of variation is:

CV = [Standard Deviation / Mean ] × 100

Features:

• Relative Measure: Unlike standard deviation, which depends on the units of measurement, CV is
unitless, allowing comparison across datasets.

• Expressed as Percentage: Since it is multiplied by 100, CV provides an intuitive understanding of

variability relative to the mean.

Applications:

8|Page
 1. Comparison Across Datasets: It is widely used in fields like finance, biology, and quality control to
compare data variability irrespective of scales.

• For instance, in finance, CV helps assess the risk (volatility) of investments relative to their
expected returns.

2. Assessing Consistency: A lower CV indicates greater consistency or homogeneity, while a higher CV

suggests greater variability or heterogeneity.

Limitations:

1. CV is only meaningful when the mean is positive and non-zero, as dividing by zero or negative means can
distort results.

2. It is less reliable for datasets with negative or near-zero values as it can exaggerate the variability.

Skewness
Skewness and Types

Skewness, in statistics, is the degree of distortion from the symmetrical bell curve, or normal distribution, in a
set of data. Skewness can be negative, positive, zero or undefined. A normal distribution has a skew of zero,
while a lognormal distribution, for example, would exhibit some degree of right-skew.

The three probability distributions depicted below depict increasing levels of right (or positive) skewness.
Distributions can also be left (negative) skewed. Skewness is used along with kurtosis to better judge the
likelihood of events falling in the tails of a probability distribution.

Right Skewness

9|Page
Key Takeaways

• Skewness, in statistics, is the degree of distortion from the symmetrical bell curve in a probability
distribution.

• Distributions can exhibit right (positive) skewness or left (negative) skewness to varying degree.

• Investors note skewness when judging a return distribution because it, like kurtosis, considers the
extremes of the data set rather than focusing solely on the average.

Broadly speaking, there are two types of skewness: They are

(1) Positive skewness

(2) Negative skewness

Positive skewness

A series is said to have positive skewness when the following characteristics are noticed:

• Mean > Median > Mode.

• The right tail of the curve is longer than its left tail, when the data are plotted through a histogram, or a
frequency polygon.

• The formula of Skewness and its coefficient give positive figures.

Negative skewness

A series is said to have negative skewness when the following characteristics are noticed:

• Mode> Median > Mode.

• The left tail of the curve is longer than the right tail, when the data are plotted through a histogram, or a
frequency polygon.

• The formula of skewness and its coefficient give negative figures.

Thus, a statistical distribution may be three types viz.

• Symmetric

• Positively skewed

• Negatively skewed

Test of Skewness Its Importance:

• Distribution Shape: Skewness helps in understanding the shape of the data distribution. A skewness of
zero indicates a symmetric distribution, while positive skewness means that the tail on the right side is
longer or fatter than the left side, and negative skewness means the opposite.

• Data Analysis and Interpretation: Recognizing skewness helps in interpreting data more accurately.
For instance, if a dataset is positively skewed, it may indicate that there are a few unusually high values
compared to the rest. This can affect statistical measures like the mean, median, and mode.

• Model Assumptions: Many statistical models, such as linear regression, assume normally distributed
errors (which implies zero skewness). Detecting skewness can help in choosing the appropriate
transformations or alternative models to better fit the data.

10 | P a g e
 • Robust Statistics: Skewness can inform the choice of statistical methods. For example, if data are
skewed, robust statistical methods or non-parametric tests may be preferred over traditional parametric
tests which assume normality.

• Financial and Risk Analysis: In finance, skewness is used to understand the risk and return profiles of
investments. Positive skewness in investment returns may indicate the potential for high gains, though
possibly with high risk, while negative skewness might indicate the risk of significant losses.

• Quality Control: In quality control and process management, skewness can help in identifying issues
with process distributions. For example, if the distribution of a product measurement is skewed, it might
indicate problems with the production process or quality control mechanisms.

Kurtosis
Kurtosis, Graph, Uses

Kurtosis is a statistical metric that quantifies the heaviness of the tails in a distribution compared to those of a
normal distribution. It highlights whether a dataset contains extreme values in its tails.

Together with skewness, kurtosis provides critical insights into data distribution. While skewness measures the
symmetry of a distribution, kurtosis focuses on the prevalence of outliers in the tails. Both are essential for
understanding distribution characteristics but capture different aspects of data behavior.

In modern finance, kurtosis plays a significant role in risk analysis. Higher kurtosis indicates a greater likelihood
of extreme returns, signaling higher risk and potential for large gains or losses in investments. Conversely, lower
kurtosis suggests more stable returns with reduced chances of outliers, indicating moderate risk.

Updated financial models often consider kurtosis for stress testing portfolios and assessing market behavior, as
extreme events are increasingly relevant in today’s volatile markets. A high kurtosis could signal vulnerability to
tail-risk events, such as market crashes or rare profit spikes, emphasizing its importance in predictive analytics
and risk management.

Excess Kurtosis: An excess kurtosis is a metric that compares the kurtosis of a distribution against the kurtosis
of a normal distribution. The kurtosis of a normal distribution equals 3. Therefore, the excess kurtosis is found
using the formula below:

Excess Kurtosis = Kurtosis – 3

Types of Kurtosis

The types of kurtosis are determined by the excess kurtosis of a particular distribution. The excess kurtosis can
take positive or negative values as well, as values close to zero.

1. Mesokurtic: Data that follows a mesokurtic distribution shows an excess kurtosis of zero or close to zero. It
means that if the data follows a normal distribution, it follows a mesokurtic distribution.

11 | P a g e
2. Leptokurtic

Leptokurtic indicates a positive excess kurtosis distribution. The leptokurtic distribution shows heavy tails on
either side, indicating the large outliers. In finance, a leptokurtic distribution shows that the investment returns
may be prone to extreme values on either side. Therefore, an investment whose returns follow a leptokurtic
distribution is considered to be risky.

3. Platykurtic

A platykurtic distribution shows a negative excess kurtosis. The kurtosis reveals a distribution with flat tails. The
flat tails indicate the small outliers in a distribution. In the finance context, the platykurtic distribution of
the investment returns is desirable for investors because there is a small probability that the investment would
experience extreme returns.

Uses of Kurtosis:

1. Risk Analysis in Finance

• Kurtosis helps in understanding the likelihood of extreme outcomes in financial returns.

• High kurtosis indicates a greater probability of extreme losses or gains, critical for risk-sensitive
investments.

• Portfolio managers and analysts use kurtosis to stress-test portfolios against market crashes or rare
profit spikes.

2. Quality Control

• In manufacturing, kurtosis aids in detecting variations in production processes.

• Low kurtosis suggests consistent production quality, while high kurtosis may point to sporadic defects.

• It ensures product reliability by identifying outliers that could compromise quality standards.

3. Decision-Making in Economics

• Economists use kurtosis to evaluate income or wealth distributions.

12 | P a g e
 • A high kurtosis indicates inequality with extreme wealth or poverty cases, influencing policies on
redistribution.

• It highlights the presence of economic shocks or anomalies in datasets.

4. Model Evaluation in Data Science

• In predictive modeling, kurtosis helps assess model accuracy by identifying outliers.

• High kurtosis datasets may require special preprocessing techniques to handle extreme values.

• It ensures robust model performance, especially in machine learning applications.

5. Healthcare and Clinical Research

• Researchers use kurtosis to identify anomalies in medical data, such as rare disease incidences or
treatment responses.

• It aids in improving patient outcomes by emphasizing extreme cases that might require targeted
interventions.

6. Actuarial Science and Insurance

• Actuaries rely on kurtosis to predict the likelihood of rare, high-cost events, such as natural disasters or
accidents.

• High kurtosis highlights the need for reserves to cover tail-risk events, ensuring financial stability for
insurers.

13 | P a g e
 Unit – II
Time Series Analysis Concept, Additive and Multiplicative Models
Time Series Analysis, Concept, Additive and Multiplicative Models

Time Series analysis is a statistical technique used to analyze data points collected or recorded at specific time
intervals. It focuses on identifying patterns, trends, seasonal variations, and cyclical behaviors within the data
over time. This method is essential for understanding historical data and making predictions or forecasts about
future values. Common components of time series include trend (long-term direction), seasonality (repeating
patterns at regular intervals), and random noise (unexplained fluctuations). Time series analysis is widely
applied in fields like finance, economics, weather forecasting, and inventory management, aiding in decision-
making and strategic planning through data-driven insights.

Components of a Time Series: Time Series typically consists of four main components, each representing
different patterns in the data. These components help in understanding the underlying structure and behavior of
the data over time. The four primary components are:

1. Trend (T)

The long-term movement or direction in the data, either upward, downward, or flat.

• Example: In sales data, a consistent increase over several years indicates an upward trend.

• Importance: Helps in understanding the general direction of the dataset over time.

2. Seasonality (S)

Repeating short-term patterns or fluctuations within a fixed period, such as daily, weekly, monthly, or yearly
cycles.

• Example: Retail sales often peak during the holiday season every year.

• Importance: Identifies periodic fluctuations that repeat at regular intervals, allowing businesses to plan
for seasonal changes.

3. Cyclic (C)

Long-term fluctuations that are irregular and occur over periods longer than a year. These cycles do not have a
fixed period and may be influenced by economic or other factors.

• Example: Economic recessions and booms are cyclical but do not follow a predictable pattern.

• Importance: Helps in understanding the impact of long-term macroeconomic factors on the data.

4. Random (Noise) (R)

The unpredictable, irregular component of the data that cannot be explained by trends, seasonality, or cycles. It
results from random events or irregular occurrences.

• Example: Sudden natural disasters, strikes, or other unplanned events.

• Importance: Identifying and accounting for noise is essential for more accurate forecasting and
analysis.

Multiplicative Models

Multiplicative models in time series analysis are used when the components of a time series (trend,
seasonality, cyclical, and random) interact in such a way that their combined effect is best represented by

14 | P a g e
multiplication rather than addition. In this model, each component of the time series is multiplied together to
explain the behavior of the data. It assumes that the effect of the components on the time series is proportional.

Mathematical Representation:

The multiplicative model can be expressed as: Yt = Tt × St × Ct × Rt

Where:

• Yt = Observed value at time t

• Tt = Trend component at time t

• St = Seasonal component at time t

• Ct = Cyclical component at time t

• Rt = Random noise at time t

When to Use the Multiplicative Model:

• Proportional Changes: The multiplicative model is suitable when the seasonal or cyclical effects
increase or decrease in proportion to the trend. For example, if sales are doubling each year and the
seasonal effect also doubles, a multiplicative model would be more appropriate than an additive one.

• Data with a Non-Constant Variability: When the variability of the data changes over time, typically as
the trend increases, this model becomes more suitable because it captures proportional variations
better.

Features:

• Proportional Relationships: Each component in the time series interacts multiplicatively, meaning the
combined effect of the components changes as the values of the components themselves change.

• Seasonal Influence: In a multiplicative model, seasonal fluctuations are expressed as a percentage of

the trend, and they amplify or reduce based on the magnitude of the trend.

• Handling Exponential Growth: The model is ideal for data where the trend is growing or shrinking
exponentially, as it allows the components to grow in proportion to each other.

Example: Consider a business where sales grow over time (trend), and there is a regular seasonal effect (e.g.,
higher sales in December). If the trend is doubling sales every year, and the seasonal effect is 1.5 times higher in
December, the multiplicative model would reflect the compounding effect of these components on the overall
sales data.

Additive Models

Additive models in time series analysis are used when the components of a time series (trend, seasonality,
cyclicity, and random noise) are assumed to influence the observed data independently and additively. This
means that the overall value of the time series is the sum of the individual components, without any interaction
between them.

Mathematical Representation:

The additive model can be expressed as: Yt = Tt + St + Ct + Rt

Where:

• Yt = Observed value at time t

• Tt = Trend component at time t

15 | P a g e
 • St = Seasonal component at time t

• Ct = Cyclical component at time t

• Rt = Random noise at time t

When to Use the Additive Model:

• Constant Variability: The additive model is appropriate when the magnitude of seasonal and cyclical
fluctuations remains relatively constant over time, regardless of the trend.

• Stable Data: It is suitable for datasets where the components of the series do not change in proportion
to the level of the trend. This is typical for data where variability or seasonality remains roughly the same
in magnitude over time.

Key Features:

• Independent Components: Each component (trend, seasonality, cyclicality, and random noise)
contributes separately to the overall time series. There is no interaction or proportionality between them.

• Constant Seasonal Effect: In an additive model, the seasonal fluctuations are the same, irrespective of
the underlying trend. For example, if sales increase by 20 units every winter, the seasonal effect would
always add 20 units to the sales data, regardless of the overall growth trend.

• Linear Growth: Trend component in an additive model represents a constant increase or decrease over
time, suitable for data that grows linearly rather than exponentially.

Example: Consider a dataset for monthly sales where the trend increases by a fixed number each year, and
seasonal fluctuations are consistent every year (e.g., higher sales in December). If sales increase by 100 units
per year and the seasonal effect is an additional 20 units in December, the additive model would assume that
the total sales in December are the sum of the trend (100) and the seasonal effect (20).

Components of Time Series

Components of time series Viz. Secular trend, Cyclical, Seasonal and irregular variations

When quantitative data are arranged in the order of their occurrence, the resulting statistical series is called a
time series. The quantitative values are usually recorded over equal time interval daily, weekly, monthly,
quarterly, half yearly, yearly, or any other time measure. Monthly statistics of Industrial Production in India,
Annual birth-rate figures for the entire world, yield on ordinary shares, weekly wholesale price of rice, daily
records of tea sales or census data are some of the examples of time series. Each has a common characteristic
of recording magnitudes that vary with passage of time.

Time series are influenced by a variety of forces. Some are continuously effective other make themselves felt at
recurring time intervals, and still others are non-recurring or random in nature. Therefore, the first task is to break
down the data and study each of these influences in isolation. This is known as decomposition of the time series.
It enables us to understand fully the nature of the forces at work. We can then analyse their combined
interactions. Such a study is known as time-series analysis.

Components of time series

A time series consists of the following four components or elements:

1. Basic or Secular or Long-time trend;

2. Seasonal variations;

16 | P a g e
 3. Business cycles or cyclical movement; and

4. Erratic or Irregular fluctuations.

These components provide a basis for the explanation of the past behaviour. They help us to predict the future
behaviour. The major tendency of each component or constituent is largely due to casual factors. Therefore a
brief description of the components and the causal factors associated with each component should be given
before proceeding further.

1. Basic or secular or long-time trend: Basic trend underlines the tendency to grow or decline over a
period of years. It is the movement that the series would have taken, had there been no seasonal, cyclical
or erratic factors. It is the effect of such factors which are more or less constant for a long time or which
change very gradually and slowly. Such factors are gradual growth in population, tastes and habits or the
effect on industrial output due to improved methods. Increase in production of automobiles and a
gradual decrease in production of food grains are examples of increasing and decreasing secular trend.

All basic trends are not of the same nature. Sometimes the predominating tendency will be a constant
amount of growth. This type of trend movement takes the form of a straight line when the trend values
are plotted on a graph paper. Sometimes the trend will be constant percentage increase or decrease.
This type takes the form of a straight line when the trend values are plotted on a semi-logarithmic chart.
Other types of trend encountered are “logistic”, “S-curyes”, etc. Properly recognising and accurately
measuring basic trends is one of the most important problems in time series analysis. Trend values are
used as the base from which other three movements are measured. Therefore, any inaccuracy in its
measurement may vitiate the entire work. Fortunately, the causal elements controlling trend growth are
relatively stable. Trends do not commonly change their nature quickly and without warning. It is therefore
reasonable to assume that a representative trend, which has characterized the data for a past period, is
prevailing at present, and that it may be projected into the future for a year or so.

2. Seasonal Variations: The two principal factors liable for seasonal changes are the climate or weather
and customs. Since, the growth of all vegetation depends upon temperature and moisture, agricultural
activity is confined largely to warm weather in the temperate zones and to the rainy or post-rainy season
in the torried zone (tropical countries or sub-tropical countries like India). Winter and dry season make
farming a highly seasonal business. This high irregularity of month to month agricultural production
determines largely all harvesting, marketing, canning, preserving, storing, financing, and pricing of farm
products. Manufacturers, bankers and merchants who deal with farmers find their business taking on
the same seasonal pattern which characterize the agriculture of their area.
The second cause of seasonal variation is custom, education or tradition. Such traditional days as
Diwali, Christmas. Id etc., product marked variations in business activity, travel, sales, gifts, finance,
accident, and vacationing.

The successful operation of any business requires that its seasonal variations be known, measured and
exploited fully. Frequently, the purchase of seasonal item is made from six months to a year in advance.
Departments with opposite seasonal changes are frequently combined in the same firm to avoid dull
seasons and to keep sales or production up during the entire year. Seasonal variations are measured as
a percentage of the trend rather than in absolute quantities. The seasonal index for any month (week,
quarter etc.) may be defined as the ratio of the normally expected value (excluding the business cycle
and erratic movements) to the corresponding trend value. When cyclical movement and erratic
fluctuations are absent in a lime series, such a series is called normal. Normal values thus are consisting
of trend and seasonal components. Thus, when normal values are divided by the corresponding trend
values, we obtain seasonal component of time series.

3. Business Cycle: Because of the persistent tendency for business to prosper, decline, stagnate
recover; and prosper again, the third characteristic movement in economic time series is called the
business cycle. The business cycle does not recur regularly like seasonal movement, but moves in

17 | P a g e
 response to causes which develop intermittently out of complex combinations of economic and other
considerations. When the business of a country or a community is above or below normal, the excess
deficiency is usually attributed to the business cycle. Its measurement becomes a process of contrast
occurrences with a normal estimate arrived at by combining the calculated trend and seasonal
movements. The measurement of the variations from normal may be made in terms of actual quantities
or it may be made in such terms as percentage deviations, which is generally more satisfactory method
as it places the measure of cyclical tendencies on comparable base throughout the entire period under
analysis.

4. Erratic or Irregular Component: These movements are exceedingly difficult to dissociate

quantitatively from the business cycle. Their causes are such irregular and unpredictable happenings
such as wars, droughts, floods, fires, pestilence, fads and fashions which operate as spurs or deterrents
upon the progress of the cycle. Examples such movements are: high activity in middle forties due to
erratic effects of 2nd world war, depression of thirties throughout the world, export boom associated
with Korean War in 1950.

The common denominator of every random factor it that does not come about as a result of the ordinary
operation of the business system and does not recur in any meaningful manner.

Mathematical Statement of the Composition of Time Series

A time series may not be affected by all type of variations. Some of these types of variations may affect a few
time series, while the other series may be effected by all of them. Hence, in analysing time series, these effects
are isolated. In classical time series analysis it is assumed that any given observation is made up of trend,
seasonal, cyclical and irregular movements and these four components have multiplicative relationship.
Symbolically:

O=T×S×C×I

where O refers to original data,

T refers to trend.

S refers to seasonal variations,

C refers to cyclical variations and

I refers lo irregular variations.

This is the most commonly used model in the decomposition of time series.
There is another model called Additive model in which a particular observation in a time series is the sum of
these four components.

O=T+S+C+I

Trend Analysis:
Applications in business Decision Making
Applications in business Decision Making

The Least Squares Method (LSM) is reliable and prevalent means to solve prediction problems in applied
research and in econometrics particularly. It is used in the case when the function is represented by its
observations. Commonly used statistical form of LSM is called Regression Analysis (RA). It is necessary to say,
that RA is only statistical shape for representing the link between the components in observations. So using RA

18 | P a g e
terminology of LSM for solution of function estimating problem, and correspondingly, – prediction problem, is
only the form for problem discussing.

It is opportune to note, that the LSM is equivalent to Maximum Likelihood Method for classic normal regression.
This method is widely used in econometric problems. The development of technical capabilities of LSM for
solution of optimization and predictive application tasks is proposed. Some examples of the least squares
method for macroeconomic models parameters identification are given in. Linear regression (LA) within RA has
the advantage of having a closed form solution of parameter estimation

Problem and prediction problem. Real valued functions of vector argument are the object of investigation in RA
in general and in LA in particular. Such suppositions are due to technical capabilities of technique for solving
optimization problems in LSM. This technique is in the essence an investigation of extremum necessary
conditions. This remark is entirely true for yet another widely used assumption, namely, full column rank
assumption for appropriate matrix, which ensure uniqueness of parameter estimation. It’s interesting that
another technique: Moore – Penrose pseudo inverse (M-Ppi) provides a comprehensive study and solution of
parameter estimation problem.

And the remark in conclusion. Obvious advantage of matrixes LSM, besides the explicit closed estimation form,
is the fact that matrixes observations preserve relationships between the characteristics of phenomenon under
consideration. Examples of matrix least square method in macroeconomic and business problems with
different types of relations between input and output data and different degree data discretization are given in.

Index Numbers Meaning, Types of index numbers, uses of index

numbers
Index Numbers: Definition, Characteristics, Uses, Types

Index Number is a statistical tool used to measure the relative change in a variable or a group of related
variables over time, across regions, or between different conditions. It simplifies comparison by converting
values into a standardized form. Index numbers are typically used to track changes in economic variables like
prices, quantities, and values, and help in measuring inflation, production levels, or economic growth.

The formula for an index number is typically:

Index Number = [Value in Current Period / Value in Base Period] × 100

Features of Index Numbers:

(i) Index numbers are a special type of average.

Whereas mean, median and mode measure the absolute changes and are used to compare only those series
which are expressed in the same units, the technique of index numbers is used to measure the relative changes
in the level of a phenomenon where the measurement of absolute change is not possible and the series are
expressed in different types of items.

(ii) Index numbers are meant to study the changes in the effects of such factors which cannot be measured
directly.

For example, the general price level is an imaginary concept and is not capable of direct measurement. But,
through the technique of index numbers, it is possible to have an idea of relative changes in the general level of
prices by measuring relative changes in the price level of different commodities.

(iii) The technique of index numbers measures changes in one variable or group of related variables.

19 | P a g e
For example, one variable can be the price of wheat, and group of variables can be the price of sugar, the price
of milk and the price of rice.

(iv) The technique of index numbers is used to compare the levels of a phenomenon on a certain date with its
level on some previous date (e.g., the price level in 1980 as compared to that in 1960 taken as the base year) or
the levels of a phenomenon at different places on the same date (e.g., the price level in India in 1980 in
comparison with that in other countries in 1980).

Uses of Index numbers:

• Index numbers are used in the fields of commerce, meteorology, labour, industry, etc.

• Index numbers measure fluctuations during intervals of time, group differences of geographical position
of degree, etc.

• They are used to compare the total variations in the prices of different commodities in which the unit of
measurements differs with time and price, etc.

• They measure the purchasing power of money.

• They are helpful in forecasting future economic trends.

• They are used in studying the difference between the comparable categories of animals, people or items.

• Index numbers of industrial production are used to measure the changes in the level of industrial
production in the country.

• Index numbers of import prices and export prices are used to measure the changes in the trade of a
country.

• Index numbers are used to measure seasonal variations and cyclical variations in a time series.

A collection of index numbers for different years, locations, etc., is sometimes called an index series.

• Simple Index Number: A simple index number is a number that measures a relative change in a single
variable with respect to a base.

• Composite Index Number: A composite index number is a number that measures an average relative
changes in a group of relative variables with respect to a base.

Types of Index Numbers:

The following types of index numbers are usually used: price index numbers and quantity index numbers.

1. Price Relative: A price relative measures the change in the price of a commodity or group of commodities
over time. It reflects how the price of a product has changed compared to its price in the base period.

• Formula:

Price Relative = [Price in Current Period / Price in Base Period] × 100

2. Quantity Relative: A quantity relative measures the change in the quantity or volume of goods produced or
consumed. It shows how much more or less of a commodity is available compared to a base period.

• Formula:

Quantity Relative = [Quantity in Current Period / Quantity in Base Period] × 100

3. Value Relative: Value relative measures the change in the value of a commodity or a group of commodities,
considering both price and quantity changes. It combines the effects of price and quantity variations on the total
value.

20 | P a g e
 • Formula:

Value Relative = [Value in Current Period / Value in Base Period] × 100

Construction of Price, Quantity and Volume indices

Price, Quantity and Value Indices

Price index (PI)

A price index (PI) is a measure of how prices change over a period of time, or in other words it is a way to measure
inflation. There are multiple methods on how to calculate the inflation (or deflation), in this guide we will take a
look at a couple of methods on how to do so. Inflation is one of the core metrics monitored by the FED in order
to set interest rates.

The general formula for the price index is the following: PI1,2 = f(P1,P2,X)

Where:

• PI1,2: Some PI that measures the change in price from period 1 to period 2

• P1: Price of goods in period 1

• P2: Price of goods in period 2

• X: Weights (the weights are used in conjunction with the prices)

Quantity index numbers

Quantity index numbers measure the change in the quantity or volume of goods sold, consumed or produced
during a given time period. Hence it is a measure of relative changes over a period of time in the quantities of a
particular set of goods.

Just like price index numbers and value index numbers, there are also two types of quantity index numbers,
namely

• Unweighted Quantity Indices

• Weighted Quantity Indices

Let us take a look at the various methods, formulas, and examples of both these types of quantity index numbers.

Value Index Number

The value index number compares the value of a commodity in the current year, with its value in the base year.
What is the value of a commodity? It is nothing but the product of the price of the commodity and the quantity.
So the value index number is the sum of the value of the commodity of the current year divided by the sum of its
value in the chosen base year. The formula is as follows,

v01 = (∑p1q1/∑p0q0) × 100

or alternatively, v01 = (∑V1/∑V0 )× 100

In this case of a value index number, we do not apply any weights. Since these are considered to be inherent in
the value of a commodity. Thus we can say that a value index number is an aggregate of values.

The value index number is not a very popular statistical tool. Price and quantity index numbers give a clearer
picture of the economy for study and analysis. They even help in the formulation and implementation of
economic policies.

21 | P a g e
 Chain Base Methods
Chain Index Numbers

In this method, there is no fixed base period; the year immediately preceding the one for which the price index
has to be calculated is assumed as the base year. Thus, for the year 1994 the base year would be 1993, for 1993
it would be 1992, for 1992 it would be 1991, and so on. In this way there is no fixed base and it keeps on changing.

The chief advantage of this method is that the price relatives of a year can be compared with the price levels of
the immediately preceding year. Businesses mostly interested in comparing this time period rather than
comparing rates related to the distant past will utilize this method.

Another advantage of the chain base method is that it is possible to include new items in an index number or to
delete old times which are no longer important. This is not possible with the fixed base method. But the chain
base method has the drawback that comparisons cannot be made over a long period.

In chain base,

Link relative of current years = (Price in the Current Year/Price in the preceding Year×100
OR

Pn−1,n= (Pn/Pn−1)×100

Example:
Find the index numbers for the following data taking 1980 as the base year.

Years 19741974 19751975 19761976 19771977 19781978 19791979

Price 1818 2121 2525 2323 2828 3030

Solution:

Year Price Link Relatives Chain Indices

=Pn/Pn−1×100

19741974 1818 1818×100=1001818×100=100 100100

19751975 2121 2118×100=116.672118×100=116.67 100×116.67100=116.67100×116.67100=116.67

19761976 2525 2521×100=119.052521×100=119.05 116.67×119.05100=138.9116.67×119.05100=138.9

19771977 2323 2325×100=922325×100=92 138.9×92100=127.79138.9×92100=127.79

22 | P a g e
 19781978 2828 2823×100=121.742823×100=121.74 127.79×121.74100=155.57127.79×121.74100=155.57

19791979 3030 3028×100=107.143028×100=107.14 155.57×107.17100=166.68155.57×107.17100=166.68

Selection of a suitable average

There are different averages which can be used in averaging the price relatives or link relatives of different
commodities. Experts have suggested that the geometric mean should be calculated to average these relatives.
But as the calculation of the geometric mean is difficult, it is mostly avoided and the arithmetic mean is
commonly used. In some cases, the median is used to remove the effect of wild observations.

Selection of suitable weights

In the calculation of price index numbers all commodities are not of equal importance. In order to give them due
importance, commodities are given due weights. Weights are of two kinds: (a) Implicit weights and (b) explicit
weights.

Implicit weights are not explicitly assigned to any commodity, but the commodity to which greater importance
is attached and is repeated a number of times. A number of varieties of such commodities are included in the
index number as separate items. Thus, if an index number wheat is to receive a weight of 3 and rice a weight of
2, three varieties of wheat and two varieties of rice included in these method weights are not apparent, but items
are implicitly weighted.

Explicit weights are explicitly assigned to commodities. Only one variety of the commodity is included in the
construction of the index number but its price relative is multiplied by the figure of weight assigned to it. Explicit
weights are decided on a logical basis. For example, if wheat and rice are to be weighted in accordance with the
value of their net output and if the ratio of their net output is 5:2, wheat would receive a weight of five and rice of
two.

Sometimes the quantities which are consumed are used as weights. These are called quantity weights. The
amount spent on different commodities can also be used as their weights. These are called the value weights.

23 | P a g e
 Unit – III
Correlation Analysis: Rank Method & Karl Pearson’s Coefficient of
Correlation
Measurement of Correlation: Karl Pearson’s Method, Spearman Rank Correlation

Karl Pearson’s Coefficient of Correlation is widely used mathematical method wherein the numerical
expression is used to calculate the degree and direction of the relationship between linear related variables.

Pearson’s method, popularly known as a Pearsonian Coefficient of Correlation, is the most extensively used
quantitative methods in practice. The coefficient of correlation is denoted by “r”.

If the relationship between two variables X and Y is to be ascertained, then the following formula is used:

Properties of Coefficient of Correlation

• The value of the coefficient of correlation (r) always lies between±1. Such as:

r=+1, perfect positive correlation

r=-1, perfect negative correlation

r=0, no correlation

The coefficient of correlation is independent of the origin and scale. By origin, it means subtracting any non-zero
constant from the given value of X and Y the value of “r” remains unchanged. By scale it means, there is no effect
on the value of “r” if the value of X and Y is divided or multiplied by any constant.

The coefficient of correlation is a geometric mean of two regression coefficient. Symbolically it is represented
as:

• The coefficient of correlation is “zero” when the variables X and Y are independent. But, however, the
converse is not true.

Assumptions of Karl Pearson’s Coefficient of Correlation:

1. The relationship between the variables is “Linear”, which means when the two variables are plotted, a straight
line is formed by the points plotted.

2. There are a large number of independent causes that affect the variables under study so as to form a Normal
Distribution. Such as, variables like price, demand, supply, etc. are affected by such factors that the normal
distribution is formed.
24 | P a g e
3. The variables are independent of each other.

Note: The coefficient of correlation measures not only the magnitude of correlation but also tells the direction.
Such as, r = -0.67, which shows correlation is negative because the sign is “-“ and the magnitude is 0.67.

Spearman Rank CORRELATION:

Spearman rank correlation is a non-parametric test that is used to measure the degree of association between
two variables. The Spearman rank correlation test does not carry any assumptions about the distribution of the
data and is the appropriate correlation analysis when the variables are measured on a scale that is at least
ordinal.

The Spearman correlation between two variables is equal to the Pearson correlation between the rank values of
those two variables; while Pearson’s correlation assesses linear relationships, Spearman’s correlation
assesses monotonic relationships (whether linear or not). If there are no repeated data values, a perfect
Spearman correlation of +1 or −1 occurs when each of the variables is a perfect monotone function of the other.

Intuitively, the Spearman correlation between two variables will be high when observations have a similar (or
identical for a correlation of 1) rank (i.e. relative position label of the observations within the variable: 1st, 2nd,
3rd, etc.) between the two variables, and low when observations have a dissimilar (or fully opposed for a
correlation of −1) rank between the two variables.

The following formula is used to calculate the Spearman rank correlation:

ρ = Spearman rank correlation

di = the difference between the ranks of corresponding variables

n = number of observations

Assumptions

The assumptions of the Spearman correlation are that data must be at least ordinal and the scores on one
variable must be monotonically related to the other variable.

Properties of Correlation
Properties of Correlation co-efficient

The following are the main properties of correlation.

1. Coefficient of Correlation lies between -1 and +1:

The coefficient of correlation cannot take value less than -1 or more than one +1. Symbolically,

-1<=r<= + 1 or | r | <1.

2. Coefficients of Correlation are independent of Change of Origin:

This property reveals that if we subtract any constant from all the values of X and Y, it will not affect the
coefficient of correlation.

3. Coefficients of Correlation possess the property of symmetry:

The degree of relationship between two variables is symmetric as shown below:

25 | P a g e
 4. Coefficient of Correlation is independent of Change of Scale:

This property reveals that if we divide or multiply all the values of X and Y, it will not affect the coefficient of
correlation.

5. Co-efficient of correlation measures only linear correlation between X and Y.

6. If two variables X and Y are independent, coefficient of correlation between them will be zero.

Karl Pearson’s Coefficient of Correlation is widely used mathematical method wherein the numerical
expression is used to calculate the degree and direction of the relationship between linear related variables.

Pearson’s method, popularly known as a Pearsonian Coefficient of Correlation, is the most extensively used
quantitative methods in practice. The coefficient of correlation is denoted by “r”.

If the relationship between two variables X and Y is to be ascertained, then the following formula is used:

Properties of Coefficient of Correlation

• The value of the coefficient of correlation (r) always lies between±1. Such as: r=+1, perfect positive
correlation r=-1, perfect negative correlation r=0, no correlation

• The coefficient of correlation is independent of the origin and scale. By origin, it means subtracting any
non-zero constant from the given value of X and Y the value of “r” remains unchanged. By scale it means,
there is no effect on the value of “r” if the value of X and Y is divided or multiplied by any constant.

• The coefficient of correlation is a geometric mean of two regression coefficient. Symbolically it is

represented as:

• The coefficient of correlation is “zero” when the variables X and Y are independent. But, however, the
converse is not true.

Assumptions of Karl Pearson’s Coefficient of Correlation

1. The relationship between the variables is “Linear”, which means when the two variables are plotted, a
straight line is formed by the points plotted.

2. There are a large number of independent causes that affect the variables under study so as to form
a Normal Distribution. Such as, variables like price, demand, supply, etc. are affected by such factors
that the normal distribution is formed.

3. The variables are independent of each other.

26 | P a g e
Note: The coefficient of correlation measures not only the magnitude of correlation but also tells the direction.
Such as, r = -0.67, which shows correlation is negative because the sign is “-“ and the magnitude is 0.67.

Regression Analysis: Fitting of a Regression Line and Interpretation of

Results
Regression Meaning, Assumption, Regression Line

Regression is a statistical measurement used in finance, investing and other disciplines that attempts to
determine the strength of the relationship between one dependent variable (usually denoted by Y) and a series
of other changing variables (known as independent variables).

Regression helps investment and financial managers to value assets and understand the relationships between
variables, such as commodity prices and the stocks of businesses dealing in those commodities.

The two basic types of regression are linear regression and multiple linear regressions, although there are non-
linear regression methods for more complicated data and analysis. Linear regression uses one independent
variable to explain or predict the outcome of the dependent variable Y, while multiple regressions use two or
more independent variables to predict the outcome.

Regression can help finance and investment professionals as well as professionals in other businesses.
Regression can also help predict sales for a company based on weather, previous sales, GDP growth or other
types of conditions. The capital asset pricing model (CAPM) is an often-used regression model in finance for
pricing assets and discovering costs of capital.

The general form of each type of regression is:

• Linear regression: Y = a + bX + u

• Multiple regression: Y = a + b1X1 + b2X2 + b3X3 + … + btXt + u

Where:

Y = the variable that you are trying to predict (dependent variable).

X = the variable that you are using to predict Y (independent variable).

a = the intercept.

b = the slope.

u = the regression residual.

Regression takes a group of random variables, thought to be predicting Y, and tries to find a mathematical
relationship between them. This relationship is typically in the form of a straight line (linear regression) that best
approximates all the individual data points. In multiple regression, the separate variables are differentiated by
using numbers with subscripts.

Assumptions in Regression

• Independence: The residuals are serially independent (no autocorrelation).

• The residuals are not correlated with any of the independent (predictor) variables.

• Linearity: The relationship between the dependent variable and each of the independent variables is
linear.

• Mean of Residuals: The mean of the residuals is zero.

27 | P a g e
 • Homogeneity of Variance: The variance of the residuals at all levels of the independent variables is
constant.

• Errors in Variables: The independent (predictor) variables are measured without error.

• Model Specification: All relevant variables are included in the model. No irrelevant variables are
included in the model.

• Normality: The residuals are normally distributed. This assumption is needed for valid tests of
significance but not for estimation of the regression coefficients.

Regression Line:

The Regression Line is the line that best fits the data, such that the overall distance from the line to the points
(variable values) plotted on a graph is the smallest. In other words, a line used to minimize the squared
deviations of predictions is called as the regression line.

There are as many numbers of regression lines as variables. Suppose we take two variables, say X and Y, then
there will be two regression lines:

• Regression line of Y on X: This gives the most probable values of Y from the given values of X.

• Regression line of X on Y: This gives the most probable values of X from the given values of Y.

The algebraic expression of these regression lines is called as Regression Equations. There will be two regression
equations for the two regression lines.

The correlation between the variables depends on the distance between these two regression lines, such as the
nearer the regression lines to each other the higher is the degree of correlation, and the farther the regression
lines to each other the lesser is the degree of correlation.

The correlation is said to be either perfect positive or perfect negative when the two regression lines coincide,
i.e. only one line exists. In case, the variables are independent; then the correlation will be zero, and the lines of
regression will be at right angles, i.e. parallel to the X axis and Y axis.

28 | P a g e
Note: The regression lines cut each other at the point of average of X and Y. This means, from the point where
the lines intersect each other the perpendicular is drawn on the X axis we will get the mean value of X. Similarly,
if the horizontal line is drawn on the Y axis we will get the mean value of Y.

Properties of Regression Coefficients

Properties of Regression Coefficients

The constant ‘b’ in the regression equation (Y e = a + bX) is called as the Regression Coefficient. It determines
the slope of the line, i.e. the change in the value of Y corresponding to the unit change in X and therefore, it is
also called as a “Slope Coefficient.”

Properties of Regression Coefficient

1. The correlation coefficient is the geometric mean of two regression coefficients. Symbolically, it can be

expressed as:

2. The value of the coefficient of correlation cannot exceed unity i.e. 1. Therefore, if one of the regression
coefficients is greater than unity, the other must be less than unity.

3. The sign of both the regression coefficients will be same, i.e. they will be either positive or negative.
Thus, it is not possible that one regression coefficient is negative while the other is positive.

4. The coefficient of correlation will have the same signas that of the regression coefficients, such as if
the regression coefficients have a positive sign, then “r” will be positive and vice-versa.

5. The average value of the two regression coefficients will be greater than the value of the correlation.

Symbolically, it can be represented as

6. The regression coefficients are independent of the change of origin, but not of the scale. By origin, we
mean that there will be no effect on the regression coefficients if any constant is subtracted from the
value of X and Y. By scale, we mean that if the value of X and Y is either multiplied or divided by some
constant, then the regression coefficients will also change.

Thus, all these properties should be kept in mind while solving for the regression coefficients.

Relationship between Regression and Correlation

Correlation and regression analysis

Correlation Analysis

29 | P a g e
Correlation is a measure of association between two variables. The variables are not designated as dependent
or independent. The two most popular correlation coefficients are: Spearman’s correlation coefficient rho and
Pearson’s product-moment correlation coefficient.

When calculating a correlation coefficient for ordinal data, select Spearman’s technique. For interval or ratio-
type data, use Pearson’s technique.

The value of a correlation coefficient can vary from minus one to plus one. A minus one indicates a perfect
negative correlation, while a plus one indicates a perfect positive correlation. A correlation of zero means there
is no relationship between the two variables. When there is a negative correlation between two variables, as the
value of one variable increases, the value of the other variable decreases, and vise versa. In other words, for a
negative correlation, the variables work opposite each other. When there is a positive correlation between two
variables, as the value of one variable increases, the value of the other variable also increases. The variables
move together.

The standard error of a correlation coefficient is used to determine the confidence intervals around a true
correlation of zero. If your correlation coefficient falls outside of this range, then it is significantly different than
zero. The standard error can be calculated for interval or ratio-type data (i.e., only for Pearson’s product-moment
correlation).

The significance (probability) of the correlation coefficient is determined from the t-statistic. The probability of
the t-statistic indicates whether the observed correlation coefficient occurred by chance if the true correlation
is zero. In other words, it asks if the correlation is significantly different than zero. When the t-statistic is
calculated for Spearman’s rank-difference correlation coefficient, there must be at least 30 cases before the t-
distribution can be used to determine the probability. If there are fewer than 30 cases, you must refer to a special
table to find the probability of the correlation coefficient.

Example

A company wanted to know if there is a significant relationship between the total number of salespeople and the
total number of sales. They collect data for five months.

Variable 1 Variable 2

207 6907

180 5991

220 6810

205 6553

190 6190

Correlation coefficient = .921

Standard error of the coefficient = ..068

30 | P a g e
t-test for the significance of the coefficient = 4.100
Degrees of freedom = 3
Two-tailed probability = .0263

Another Example

Respondents to a survey were asked to judge the quality of a product on a four-point Likert scale (excellent,
good, fair, poor). They were also asked to judge the reputation of the company that made the product on a three-
point scale (good, fair, poor). Is there a significant relationship between respondents perceptions of the
company and their perceptions of quality of the product?

Since both variables are ordinal, Spearman’s method is chosen. The first variable is the rating for the quality the
product. Responses are coded as 4=excellent, 3=good, 2=fair, and 1=poor. The second variable is the perceived
reputation of the company and is coded 3=good, 2=fair, and 1=poor.

Variable 1 Variable 2

4 3

2 2

1 2

3 3

4 3

1 1

2 1

Correlation coefficient rho = .830

t-test for the significance of the coefficient = 3.332
Number of data pairs = 7

Probability must be determined from a table because of the small sample size.

Regression Analysis

Simple regression is used to examine the relationship between one dependent and one independent variable.
After performing an analysis, the regression statistics can be used to predict the dependent variable when the
independent variable is known. Regression goes beyond correlation by adding prediction capabilities.

31 | P a g e
People use regression on an intuitive level every day. In business, a well-dressed man is thought to be financially
successful. A mother knows that more sugar in her children’s diet results in higher energy levels. The ease of
waking up in the morning often depends on how late you went to bed the night before. Quantitative regression
adds precision by developing a mathematical formula that can be used for predictive purposes.

For example, a medical researcher might want to use body weight (independent variable) to predict the most
appropriate dose for a new drug (dependent variable). The purpose of running the regression is to find a formula
that fits the relationship between the two variables. Then you can use that formula to predict values for the
dependent variable when only the independent variable is known. A doctor could prescribe the proper dose
based on a person’s body weight.

The regression line (known as the least squares line) is a plot of the expected value of the dependent variable for
all values of the independent variable. Technically, it is the line that “minimizes the squared residuals”. The
regression line is the one that best fits the data on a scatterplot.

Using the regression equation, the dependent variable may be predicted from the independent variable. The
slope of the regression line (b) is defined as the rise divided by the run. The y intercept (a) is the point on the y
axis where the regression line would intercept the y axis. The slope and y intercept are incorporated into the
regression equation. The intercept is usually called the constant, and the slope is referred to as the coefficient.
Since the regression model is usually not a perfect predictor, there is also an error term in the equation.

In the regression equation, y is always the dependent variable and x is always the independent variable. Here are
three equivalent ways to mathematically describe a linear regression model.

y = intercept + (slope x) + error

y = constant + (coefficient x) + error

y = a + bx + e

The significance of the slope of the regression line is determined from the t-statistic. It is the probability that the
observed correlation coefficient occurred by chance if the true correlation is zero. Some researchers prefer to
report the F-ratio instead of the t-statistic. The F-ratio is equal to the t-statistic squared.

The t-statistic for the significance of the slope is essentially a test to determine if the regression model (equation)
is usable. If the slope is significantly different than zero, then we can use the regression model to predict the
dependent variable for any value of the independent variable.

On the other hand, take an example where the slope is zero. It has no prediction ability because for every value
of the independent variable, the prediction for the dependent variable would be the same. Knowing the value of
the independent variable would not improve our ability to predict the dependent variable. Thus, if the slope is
not significantly different than zero, don’t use the model to make predictions.

The coefficient of determination (r-squared) is the square of the correlation coefficient. Its value may vary from
zero to one. It has the advantage over the correlation coefficient in that it may be interpreted directly as the
proportion of variance in the dependent variable that can be accounted for by the regression equation. For
example, an r-squared value of .49 means that 49% of the variance in the dependent variable can be explained
by the regression equation. The other 51% is unexplained.

32 | P a g e
The standard error of the estimate for regression measures the amount of variability in the points around the
regression line. It is the standard deviation of the data points as they are distributed around the regression line.
The standard error of the estimate can be used to develop confidence intervals around a prediction.

Example

A company wants to know if there is a significant relationship between its advertising expenditures and its sales
volume. The independent variable is advertising budget and the dependent variable is sales volume. A lag time
of one month will be used because sales are expected to lag behind actual advertising expenditures. Data was
collected for a six month period. All figures are in thousands of dollars. Is there a significant relationship between
advertising budget and sales volume?

Indep. Var. Depen. Var

4.2 27.1

6.1 30.4

3.9 25.0

5.7 29.7

7.3 40.1

5.9 28.8

Model: y = 9.873 + (3.682x) + error

Standard error of the estimate = 2.637
t-test for the significance of the slope = 3.961
Degrees of freedom = 4
Two-tailed probability = .0149
r-squared = .807

You might make a statement in a report like this: A simple linear regression was performed on six months of data
to determine if there was a significant relationship between advertising expenditures and sales volume. The t-
statistic for the slope was significant at the .05 critical alpha level, t(4)=3.96, p=.015. Thus, we reject the null
hypothesis and conclude that there was a positive significant relationship between advertising expenditures and
sales volume. Furthermore, 80.7% of the variability in sales volume could be explained by advertising
expenditures.

33 | P a g e
 Unit -IV
Probability: Theory of Probability
Probability Meaning and Approaches of Probability Theory

In our day to day life the “probability” or “chance” is very commonly used term. Sometimes, we use to say
“Probably it may rain tomorrow”, “Probably Mr. X may come for taking his class today”, “Probably you are right”.
All these terms, possibility and probability convey the same meaning. But in statistics probability has certain
special connotation unlike in Layman’s view.

The theory of probability has been developed in 17th century. It has got its origin from games, tossing coins,
throwing a dice, drawing a card from a pack. In 1954 Antoine Gornband had taken an initiation and an interest
for this area.

After him many authors in statistics had tried to remodel the idea given by the former. The “probability” has
become one of the basic tools of statistics. Sometimes statistical analysis becomes paralyzed without the
theorem of probability. “Probability of a given event is defined as the expected frequency of occurrence of
the event among events of a like sort.” (Garrett)

The probability theory provides a means of getting an idea of the likelihood of occurrence of different events
resulting from a random experiment in terms of quantitative measures ranging between zero and one. The
probability is zero for an impossible event and one for an event which is certain to occur.

Approaches of Probability Theory

1. Classical Probability: The classical approach to probability is one of the oldest and simplest school of
thought. It has been originated in 18th century which explains probability concerning games of chances
such as throwing coin, dice, drawing cards etc.

The definition of probability has been given by a French mathematician named “Laplace”. According to
him probability is the ratio of the number of favourable cases among the number of equally likely cases.

Or in other words, the ratio suggested by classical approach is:

Pr. = Number of favourable cases/Number of equally likely cases

For example, if a coin is tossed, and if it is asked what is the probability of the occurrence of the head, then the
number of the favourable case = 1, the number of the equally likely cases = 2.

Pr. of head = 1/2

Symbolically it can be expressed as:

P = Pr. (A) = a/n, q = Pr. (B) or (not A) = b/n

1 – a/n = b/n = (or) a + b = 1 and also p + q = 1

p = 1 – q, and q = 1 – p and if a + b = 1 then so also a/n + b/n = 1

In this approach the probability varies from 0 to 1. When probability is zero it denotes that it is impossible to
occur.

If probability is 1 then there is certainty for occurrence, i.e. the event is bound to occur.

Example: From a bag containing 20 black and 25 white balls, a ball is drawn randomly. What is the probability
that it is black.

Pr. of a black ball = 20/45 = 4/9 = p, 25 Pr. of a white ball = 25/45 = 5/9 = q

34 | P a g e
p = 4/9 and q = 5/9 (p + q= 4/9 + 5/9= 1)

2. Relative Frequency Theory of Probability

This approach to probability is a protest against the classical approach. It indicates the fact that if n is increased
upto the ∞, we can find out the probability of p or q.

Example:

If n is ∞, then Pr. of A= a/n = .5, Pr. of B = b/n = 5

If an event occurs a times out of n its relative frequency is a/n. When n becomes ∞, is called the limit of relative
frequency.

Pr. (A) = limit a/n

where n → ∞

Pr. (B) = limit bl.t. here → ∞.

Axiomatic approach

An axiomatic approach is taken to define probability as a set function where the elements of the domain are the
sets and the elements of range are real numbers. If event A is an element in the domain of this function, P(A) is
the customary notation used to designate the corresponding element in the range.

Probability Function

A probability function p(A) is a function mapping the event space A of a random experiment into the interval [0,1]
according to the following axioms;

Axiom 1. For any event A, 0 ≤ P(A) ≤ 1

Axiom 2. P(Ω) = 1

Axiom 3. If A and B are any two mutually exclusive events then,

P(A ∪ B)) = P(A) + P(B)

As given in the third axiom the addition property of the probability can be extended to any number of events as
long as the events are mutually exclusive. If the events are not mutually exclusive then;

P(A ∪ B) = P(A) + P(B) – P(A∩B)

P(A∩B) is Φ if both the events are mutually exclusive.

If there are two types of objects among the objects of similar or other natures then the probability of one object
i.e. Pr. of A = .5, then Pr. of B = .5.

Addition and Multiplication Laws

Addition and Multiplication Theorems

Addition Theorem on probability:

If A and B are any two events then the probability of happening of at least one of the events is defined as P(AUB)
= P(A) + P(B) – P(A∩B).

Proof:

35 | P a g e
Since events are nothing but sets,

n(AUB) = n(A) + n(B) – n(A∩B).

Dividing the above equation by n(S), (where S is the sample space)

n(AUB)/ n(S) = n(A)/ n(S) + n(B)/ n(S)- n(A∩B)/ n(S)

Then by the definition of probability,

P(AUB) = P(A) + P(B) – P(A∩B).

Example:

If the probability of solving a problem by two students George and James are 1/2 and 1/3 respectively then what
is the probability of the problem to be solved.

Solution:

Let A and B be the probabilities of solving the problem by George and James respectively.

Then P(A) = 1/2 and P(B) = 1/3.

The problem will be solved if it is solved at least by one of them also.

So, we need to find P(AUB).

By addition theorem on probability, we have

P(AUB) = P(A) + P(B)- P(A∩B).

P(AUB) = 1/2 +.1/3 – 1/2 * 1/3 = 1/2 +1/3-1/6 = (3+2-1)/6 = 4/6 = 2/3

Note:

If A and B are any two mutually exclusive events then P(A∩B)=0.

Then P(AUB) = P(A)+P(B).

Multiplication Theorem on Probability

If A and B are any two events of a sample space such that P(A) ≠0 and P(B) ≠ 0, then

P(A∩B) = P(A) * P(B|A) = P(B) *P(A|B).

Example: If P(A) = 1/5 P(B|A) = 1/3 then what is P(A∩B)?

Solution: P(A∩B) = P(A) * P(B|A) = 1/5 * 1/3 = 1/15

INDEPENDENT EVENTS:

Two events A and B are said to be independent if there is no change in the happening of an event with the
happening of the other event.

i.e. Two events A and B are said to be independent if

P(A|B) = P(A) where P(B)≠0.

P(B|A) = P(B) where P(A)≠0.

i.e. Two events A and B are said to be independent if

P(A∩B) = P(A) * P(B).

Example:
36 | P a g e
While laying the pack of cards, let A be the event of drawing a diamond and B be the event of drawing an ace.

Then P(A) = 13/52 = 1/4 and P(B) = 4/52=1/13

Now, A∩B = drawing a king card from hearts.

Then P(A∩B) = 1/52

Now, P(A/B) = P(A∩B)/P(B) = (1/52)/(1/13) = 1/4 = P(A).

So, A and B are independent.

[Here, P(A∩B) = = = P(A) * P(B)]

Note:

(1) If 3 events A,B and C are independent the

P(A∩B∩C) = P(A)*P(B)*P(C).

(2) If A and B are any two events, then P(AUB) = 1-P(A’)P(B’).

Baye’s Theorem
Baye’s Theorem, Uses

Bayes’ Theorem is a mathematical formula used to calculate the probability of an event based on prior
knowledge of related conditions.

Bayes’ theorem is a way to figure out conditional probability. Conditional probability is the probability of an event
happening, given that it has some relationship to one or more other events. For example, your probability of
getting a parking space is connected to the time of day you park, where you park, and what conventions are going
on at any time. Bayes’ theorem is slightly more nuanced. In a nutshell, it gives you the actual probability of
an event given information about tests.

“Events” Are different from “tests.” For example, there is a test for liver disease, but that’s separate from
the event of actually having liver disease.

Tests are flawed:

Just because you have a positive test does not mean you actually have the disease. Many tests have a high false
positive rate. Rare events tend to have higher false positive rates than more common events. We’re not just
talking about medical tests here. For example, spam filtering can have high false positive rates. Bayes’ theorem
takes the test results and calculates your real probability that the test has identified the event.

Bayes’ Theorem (also known as Bayes’ rule) is a deceptively simple formula used to calculate conditional
probability. The Theorem was named after English mathematician Thomas Bayes (1701-1761). The formal
definition for the rule is:

In most cases, you can’t just plug numbers into an equation; You have to figure out what your “tests” and
“events” are first. For two events, A and B, Bayes’ theorem allows you to figure out p(A|B) (the probability that
event A happened, given that test B was positive) from p(B|A) (the probability that test B happened, given that
37 | P a g e
event A happened). It can be a little tricky to wrap your head around as technically you’re working backwards;
you may have to switch your tests and events around, which can get confusing. An example should clarify what
I mean by “switch the tests and events around.”

Bayes’ Theorem Example

You might be interested in finding out a patient’s probability of having liver disease if they are an alcoholic. “Being
an alcoholic” is the test (kind of like a litmus test) for liver disease.

A could mean the event “Patient has liver disease.” Past data tells you that 10% of patients entering your clinic
have liver disease. P(A) = 0.10.

B could mean the litmus test that “Patient is an alcoholic.” Five percent of the clinic’s patients are alcoholics.
P(B) = 0.05.

You might also know that among those patients diagnosed with liver disease, 7% are alcoholics. This is
your B|A: the probability that a patient is alcoholic, given that they have liver disease, is 7%.

Bayes’ theorem tells you:

P(A|B) = (0.07 * 0.1) / 0.05 = 0.14

In other words, if the patient is an alcoholic, their chances of having liver disease is 0.14 (14%). This is a large
increase from the 10% suggested by past data. But it’s still unlikely that any particular patient has liver disease.

Uses of Bayes’ Theorem:

• Medical Diagnosis: Bayes’ theorem helps in determining the likelihood of a disease based on diagnostic
test results. For instance, given a positive test result, the theorem updates the probability of a patient
having the disease, considering the test’s sensitivity, specificity, and disease prevalence. It enhances
diagnostic accuracy by incorporating prior probabilities.

• Spam Email Filtering: Email systems use Bayes’ theorem to classify emails as spam or legitimate. By
analyzing the likelihood of certain words appearing in spam emails versus non-spam ones, the theorem
predicts whether a new email is spam. This approach forms the basis of Bayesian spam filters, improving
email management efficiency.

• Decision-Making Under Uncertainty: In risk management and decision-making, Bayes’ theorem

evaluates the probabilities of various outcomes given new evidence. For example, in financial markets,
it helps assess the probability of market changes based on economic indicators, enabling better
investment strategies and minimizing risks in uncertain scenarios.

• Machine Learning and Artificial Intelligence: Bayesian models are core to machine learning
algorithms, especially in probabilistic reasoning. Applications include predictive modeling, such as
customer behavior analysis, and classification tasks, like image recognition. The theorem helps update
predictions or classifications as new data becomes available, ensuring adaptive and accurate systems.

Probability Theoretical Distributions: Concept

Probability Distribution: Binominal Poisson, Normal Distribution

Probability Distributions

Probability theory is the foundation for statistical inference. A probability distribution is a device for indicating
the values that a random variable may have. There are two categories of random variables. These are discrete
random variables and continuous random variables.

38 | P a g e
Discrete random variable

The probability distribution of a discrete random variable specifies all possible values of a discrete random
variable along with their respective probabilities.

Examples can be

• Frequency distribution

• Probability distribution (relative frequency distribution)

• Cumulative frequency

Examples of discrete probability distributions are the binomial distribution and the Poisson distribution.

Binomial Distribution

A binomial experiment is a probability experiment with the following properties.

1. Each trial can have only two outcomes which can be considered success or failure.

2. There must be a fixed number of trials.

3. The outcomes of each trial must be independent of each other.

4. The probability of success must remain the same in each trial.

The outcomes of a binomial experiment are called a binomial distribution.

Poisson Distribution

The Poisson distribution is based on the Poisson process.

1. The occurrences of the events are independent in an interval.

2. An infinite number of occurrences of the event are possible in the interval.

3. The probability of a single event in the interval is proportional to the length of the interval.

4. In an infinitely small portion of the interval, the probability of more than one occurrence of the event is
negligible.

Continuous probability distributions

A continuous variable can assume any value within a specified interval of values assumed by the variable. In a
general case, with a large number of class intervals, the frequency polygon begins to resemble a smooth curve.

A continuous probability distribution is a probability density function. The area under the smooth curve is equal
to 1 and the frequency of occurrence of values between any two points equals the total area under the curve
between the two points and the x-axis.

The Normal Distribution

The normal distribution is the most important distribution in biostatistics. It is frequently called the Gaussian
distribution. The two parameters of the normal distribution are the mean (m) and the standard deviation (s). The
graph has a familiar bell-shaped curve.

39 | P a g e
 Graph of a Normal Distribution

Characteristics of the normal distribution

1. It is symmetrical about m.

2. The mean, median and mode are all equal.

3. The total area under the curve above the x-axis is 1 square unit. Therefore 50% is to the right of m and 50% is
to the left of m.

4. Perpendiculars of:

± s contain about 68%;

±2 s contain about 95%;
±3 s contain about 99.7%
of the area under the curve.

The standard normal distribution

A normal distribution is determined by m and s. This creates a family of distributions depending on whatever the
values of m and s are. The standard normal distribution has m =0 and s =1.

Finding normal curve areas

1. The table gives areas between – and the value of.

2. Find the z value in tenths in the column at left margin and locate its row. Find the hundredths place in the
appropriate column.

3. Read the value of the area (P) from the body of the table where the row and column intersect. Note that P is
the probability that a given value of z is as large as it is in its location. Values of P are in the form of a decimal
point and four places. This constitutes a decimal percent.

Finding probabilities

We find probabilities using the table and a four-step procedure as illustrated below.

a) What is the probability that z < -1.96?

40 | P a g e
(1) Sketch a normal curve
(2) Draw a line for z = -1.96
(3) Find the area in the table
(4) The answer is the area to the left of the line P(z < -1.96) = .0250

b) What is the probability that -1.96 < z < 1.96?

(1) Sketch a normal curve

(2) Draw lines for lower z = -1.96, and upper z = 1.96
(3) Find the area in the table corresponding to each value
(4) The answer is the area between the values–subtract lower from upper P(-1.96 < z < 1.96) = .9750 – .0250 =
.9500

c) What is the probability that z > 1.96?

41 | P a g e
(1) Sketch a normal curve
(2) Draw a line for z = 1.96
(3) Find the area in the table
(4) The answer is the area to the right of the line; found by subtracting table value from 1.0000; P(z > 1.96) =1.0000
– .9750 = .0250

Applications of the Normal distribution

The normal distribution is used as a model to study many different variables. We can use the normal distribution
to answer probability questions about random variables. Some examples of variables that are normally
distributed are human height and intelligence.

Solving normal distribution application problems

In this explanation we add an additional step. Following the model of the normal distribution, a given value of x
must be converted to a z score before it can be looked up in the z table.

(1) Write the given information

(2) Sketch a normal curve
(3) Convert x to a z score
(4) Find the appropriate value(s) in the table
(5) Complete the answer

Illustrative Example: Total fingerprint ridge count in humans is approximately normally distributed with mean of
140 and standard deviation of 50. Find the probability that an individual picked at random will have a ridge count
less than 100. We follow the steps to find the solution.

(1) Write the given information

m = 140

s = 50

x = 100

(3) Convert x to a z score

(4) Find the appropriate value(s) in the table

42 | P a g e
 A value of z = -0.8 gives an area of .2119 which corresponds to the probability P (z < -0.8)

(5) Complete the answer

The probability that x is less than 100 is .2119.

Application of Binomial; Poisson and Normal distributions

Application of Binomial; Poisson and Normal distributions

The binomial distribution has its applications in experiments in probability subject to certain constraints. These
are:

1. There is a fixed number of trials – for example toss a coin 20 times.

2. The outcomes are independent and there are just two possible outcomes-in the example I will use, these
are head and tail.

3. The probability of a head plus the probability of a tail must equal 1.

4. The probability of 8 heads and 12 tails would be 20C8 x P(H)^8 x P(T)^12.

5. Now any experiment in which the outcomes are of just two kinds and whose probability combined equals
1, can be regarded as binomial.

Data from the analyses of reference samples often must be used to determine the quality of the data being
produced by laboratories that routinely make chemical analyses of environmental samples. When a laboratory
analyzes many reference samples, binomial distributions can be used in evaluating laboratory performance. The
number of standard deviations (that is, the difference between the reported value and most probable value
divided by the theoretical standard deviation) is calculated for each analysis. Individual values exceeding two
standard deviations are considered unacceptable, and a binomial distribution is used to determine if overall
performance is satisfactory or unsatisfactory. Similarly, analytical bias is examined by applying a binomial
distribution to the number of positive and negative standard deviations.

43 | P a g e
 Unit – V
Decision Making under Certainty, Uncertainty and Risk Situations
Decision-making under Certainty, Uncertainty and Risk Situations

Decision-making is a fundamental aspect of management and involves choosing between alternatives based
on various factors. The environment in which decisions are made can significantly impact the approach taken.
Decision-making can be categorized into three primary situations: certainty, uncertainty, and risk.

Decision-Making Under Certainty:

Decision-making under certainty occurs when the outcomes of each alternative are known with absolute
assurance. In this scenario, decision-makers have complete information about the consequences of their
choices, which allows them to predict outcomes accurately.

Characteristics:

• Clear Outcomes: Each alternative leads to a specific, identifiable result.

• Complete Information: Decision-makers have all relevant data and knowledge needed to make
informed choices.

• Stability: The environment is stable and predictable, meaning that conditions are unlikely to change.

Approach: When faced with certainty, decision-makers typically use rational models to evaluate alternatives.
They analyze the available options based on predetermined criteria, often employing techniques like linear
programming or optimization models. Since outcomes are known, the decision-making process is
straightforward, involving a logical assessment of the best alternative.

Example: A classic example of decision-making under certainty is a manufacturing firm deciding to produce a
certain number of units of a product based on established demand forecasts. If the demand is predictable and
consistent, the firm can confidently determine the production level needed to meet customer needs without the
risk of overproduction or underproduction.

Decision-Making Under Uncertainty

Decision-making under uncertainty occurs when decision-makers face situations where the outcomes of
alternatives are unknown, and they lack complete information. In this scenario, the likelihood of various
outcomes cannot be determined, making it challenging to predict results accurately.

Characteristics:

• Unknown Outcomes: The results of alternatives are unpredictable and not quantifiable.

• Incomplete Information: Decision-makers may lack critical data, making it difficult to evaluate options
effectively.

• Complexity: The environment is often dynamic and influenced by numerous unpredictable factors.

Approach: In uncertain situations, decision-makers may rely on subjective judgment, experience, and intuition.
Various techniques can be employed to navigate uncertainty, such as scenario planning, expert opinions, and
decision trees. These methods help decision-makers evaluate potential outcomes and identify feasible
alternatives despite the lack of certainty.

Example: A company considering entering a new market may face uncertainty regarding consumer preferences,
competitive dynamics, and regulatory challenges. Without clear information, decision-makers must rely on

44 | P a g e
market research, pilot testing, and expert insights to inform their strategy. The organization might develop
multiple scenarios to explore various possibilities and assess potential outcomes.

Decision-Making Under Risk

Decision-making under risk occurs when decision-makers have partial information about the probabilities of
different outcomes. In this scenario, the likelihood of specific results can be estimated, enabling decision-
makers to assess alternatives based on expected values.

Characteristics:

• Probabilistic Outcomes: The outcomes of alternatives are known, but their probabilities are uncertain.

• Quantifiable Risks: Decision-makers can assign probabilities to different outcomes based on historical
data or statistical analysis.

• Risk Management: Decision-makers can evaluate risks and returns associated with each alternative.

Approach: In risk situations, decision-makers typically use quantitative methods to analyze alternatives.
Techniques such as expected value analysis, decision trees, and Monte Carlo simulations can be employed to
evaluate options based on their associated risks and potential rewards. This structured approach allows
decision-makers to balance risks against potential benefits.

Example: An investment manager deciding to allocate funds to different financial assets faces a risk situation.
They can analyze historical returns and volatilities to estimate the probabilities of various outcomes. By using
expected value calculations, the manager can determine which investment portfolio offers the best risk-return
trade-off and make an informed decision based on quantitative analysis.

Summary:

• Decision-Making Under Certainty: Involves known outcomes and complete information, allowing for
straightforward, rational decision-making.

• Decision-Making Under Uncertainty: Involves unknown outcomes and incomplete information,

requiring decision-makers to rely on intuition, experience, and scenario analysis to navigate ambiguity.

• Decision-Making Under Risk: Involves probabilistic outcomes with quantifiable risks, enabling
decision-makers to utilize quantitative methods to evaluate alternatives based on expected values.

Decision Tree approach and its applications

Decision Tree Analysis

Decision Tree may be understood as the logical tree, is a range of conditions (premises) and actions
(conclusions), which are depicted as nodes and the branches of the tree which link the premises with
conclusions. It is a decision support tool, having a tree-like representation of decisions and the consequences
thereof. It uses ‘AND’ and ‘OR’ operators, to recreate the structure of if-then rules.

A decision tree is helpful in reaching the ideal decision for intricate processes, especially when the decision
problems are interconnected and chronological in nature.

A decision tree does not constitute a decision but assists in making one, by graphically representing the
material information related to the given problem, in the form of a tree. It diagrammatically depicts various
courses of action, likely outcomes, states of nature, etc, as nodes, branches or sub-branches of a horizontal
tree.

Nodes
45 | P a g e
There are two types of Nodes:

• Decision Node: Represented as square, wherein different courses of action arise from decision node in
main branches.

• Chance Node: Symbolised as a circle, at the terminal point of decision node, the chance node is
present, where they emerge as sub-branches. These depict probabilities and outcomes.

For instance: Think of a situation where a firm introduces a new product. The decision tree presented below
gives a clear idea of managerial problems.

• Key A is a decision node, wherein the decision is taken, i.e. to test the product or drop the same.

• Key B is an outcome node, which shows all possible outcomes, that can be taken. As per the given
situation, there are only two outcomes, i.e. favorable or not.

• Key C is again a decision node, that describes the market test is positive, so the firm’s management will
decide whether to go further with complete marketing or drop the product.

• Key D is one more decision node, but does not shows any choice, which depicts that if the market test is
unfavorable then the decision is to drop the product.

• Key E is again an outcome node.

The decision tree can be applied to various areas, where decisions are pending such as make or buy
decision, investment decision, marketing strategy, the introduction of a new project. The decision maker will go
for the alternative that increases the anticipated profit or the one which reduces the overall expected cost at
each decision point.

Types of Decision Tree

46 | P a g e
In a single stage decision tree, the decision maker can find only one solution, which is the best course of action,
on the basis of the information gathered. On the other hand, multi-stage decision tree involves a series of the
decision to be taken.

Decision Tree Analysis

The Decision Tree Analysis is a schematic representation of several decisions followed by different chances of
the occurrence. Simply, a tree-shaped graphical representation of decisions related to the investments and the
chance points that help to investigate the possible outcomes is called as a decision tree analysis.

The decision tree shows Decision Points, represented by squares, are the alternative actions along with the
investment outlays, that can be undertaken for the experimentation. These decisions are followed by
the chance points, represented by circles, are the uncertain points, where the outcomes are dependent on the
chance process. Thus, the probability of occurrence is assigned to each chance point.

Once the decision tree is described precisely, and the data about outcomes along with their probabilities is
gathered, the decision alternatives can be evaluated as follows:

1. Start from the extreme right-hand end of the tree and start calculating NPV for each chance points as you
proceed leftward.

47 | P a g e
 2. Once the NPVs are calculated for each chance point, evaluate the alternatives at the final stage decision
points in terms of their NPV.

3. Select the alternative which has the highest NPV and cut the branch of inferior decision alternative.
Assign value to each decision point equivalent to the NPV of the alternative selected.

4. Again, repeat the process, proceed leftward, recalculate NPV for each chance point, select the decision
alternative which has the highest NPV value and then cut the branch of the inferior decision alternative.
Assign the value to each point equivalent to the NPV of selected alternative and repeat this process again
and again until a final decision point is reached.

Thus, decision tree analysis helps the decision maker to take all the possible outcomes into the consideration
before reaching a final investment decision.

A decision tree is a decision support tool that uses a tree-like model of decisions and their possible
consequences, including chance event outcomes, resource costs, and utility. It is one way to display
an algorithm that only contains conditional control statements.

Decision trees are commonly used in operations research, specifically in decision analysis, to help identify a
strategy most likely to reach a goal, but are also a popular tool in machine learning.

Concept of Business Analytics Meaning, Types

Business Analytics Introduction, Concept, Evolution, Scope, Challenges

Business Analytics refers to the practice of using quantitative methods to derive meaningful insights from data
to inform business decisions. This field integrates skills from statistics, information technology, and business,
applying data-driven techniques to solve complex problems and improve a company’s performance. Business
analytics involves the collection, processing, and analysis of vast amounts of data to predict trends, optimize
processes, and enhance strategic planning. It encompasses various methodologies including descriptive
analytics (examining past performance), predictive analytics (forecasting future scenarios), and prescriptive
analytics (suggesting actions based on predictions). Tools commonly used in business analytics include data
visualization software, statistical analysis tools, and business intelligence platforms. This approach not only
helps businesses understand their current state but also anticipates future opportunities and challenges.

Evolution of Business Analytics:

1. Early Beginnings (1950s-1960s)

Initially, business analytics was mainly about basic data collection and processing, primarily done manually or
with the help of simple mechanical tools. The focus was on operational reporting and tracking standard financial
metrics.

2. The Advent of Computers (1970s-1980s)

With the introduction of computers and enterprise resource planning (ERP) systems, organizations began to
automate data collection and management. This era saw the development of more sophisticated data
processing techniques and the beginning of statistical analysis in business.

3. Rise of Business Intelligence (1990s)

Business intelligence (BI) platforms emerged, enabling more complex data analysis and reporting. Tools like SQL
for querying databases and OLAP (Online Analytical Processing) for multidimensional analysis became popular.
Companies started to leverage historical data to gain insights into business operations and performance.

4. Expansion into Predictive Analytics (2000s)

48 | P a g e
As computational power increased and data storage became more affordable, businesses began exploring
predictive analytics. Technologies such as machine learning started to be applied to predict future trends based
on historical data. This era also saw the rise of data mining techniques to discover patterns in large datasets.

5. Big Data and Real-Time Analytics (2010s)

The explosion of big data, fueled by the digitalization of business processes and the advent of the internet, social
media, and IoT, led to the need for tools that could handle vast and varied data sets in real time. Technologies
like Hadoop and Spark facilitated the processing of big data, while analytics shifted towards real-time insights,
enhancing decision-making speed and efficiency.

6. AI-Driven Analytics and Automation (2020s)

Today, business analytics is increasingly intertwined with artificial intelligence and machine learning,
automating complex analytical processes and enabling more accurate and nuanced insights. AI-driven analytics
supports prescriptive analytics, not only predicting future trends but also recommending actions. This
integration helps businesses optimize operations, personalize customer experiences, and innovate their
strategies proactively.

7. Future Trends

Looking forward, the evolution of business analytics is likely to focus on even more advanced uses of AI, deeper
integration of real-time analytics across all business processes, and greater emphasis on data privacy and
security. The use of analytics as a service (AaaS) and cloud-based analytics platforms are expected to grow,
making sophisticated analytics capabilities more accessible to smaller enterprises.

Scope of Business Analytics:

• Marketing: Business analytics helps in understanding customer behavior, segmentation, and

preferences, thereby enabling more targeted marketing strategies. Analytics can optimize marketing
campaigns by analyzing customer responses and effectiveness across different channels. Predictive
analytics is also used to forecast customer responses to new products or promotions.

• Sales: Sales analytics includes tracking sales trends, forecasting future sales, and optimizing pricing
strategies to maximize revenue. It also helps in identifying potential sales opportunities by analyzing
customer data and market conditions.

• Finance: In finance, business analytics is used for risk analysis, fraud detection, budgeting, and financial
forecasting. Analytics enables more precise financial decision-making by modeling cash flow scenarios
and assessing investment risks.

• Human Resources: Analytics in HR, often referred to as people analytics, involves analyzing employee
data to improve hiring practices, employee retention, and productivity. It also helps in predicting
employee turnover and optimizing workforce management based on business needs.

• Supply Chain Management: Business analytics enhances supply chain efficiency by optimizing
inventory levels, improving logistics and distribution strategies, and predicting supply chain disruptions.
It can also help in vendor performance management and cost reduction.

• Operations: Operational analytics uses data analysis to improve efficiency, quality, and performance.
It includes optimizing business processes, managing resources effectively, and reducing waste.
Analytics can also be used to enhance production planning and maintenance schedules.

• Customer Service: Analytics helps improve customer service by providing insights into customer
complaints and feedback. Predictive analytics can be used to anticipate customer issues before they
occur, while prescriptive analytics can suggest the best actions to resolve ongoing issues.

49 | P a g e
 • Product Development: By analyzing customer feedback, market trends, and competitive offerings,
business analytics aids in designing products that meet market needs more effectively. It can also
shorten product development cycles and improve innovation processes.

• Healthcare: In healthcare, analytics is used for improving patient care, managing healthcare costs, and
predicting disease outbreaks. It also plays a crucial role in medical research and population health
management.

• E-commerce: E-commerce businesses rely heavily on analytics for customer segmentation,

personalization strategies, optimizing the customer journey, and improving conversion rates.

• Banking and Financial Services: Analytics in banking involves credit scoring, risk management,
customer profitability analysis, and regulatory compliance. It helps in detecting and preventing fraud and
enhancing customer service.

Challenges of Business Analytics:

1. Data Quality and Accuracy: One of the most significant challenges is ensuring the accuracy,
completeness, and reliability of data. Poor data quality can lead to misleading analysis results,
impacting decision-making. Regular data cleansing and validation are essential to maintain the integrity
of business analytics.

2. Data Integration: Organizations often struggle with integrating data from multiple sources, including
internal systems and external datasets. Disparate data formats and structures can complicate the
aggregation process, affecting the overall effectiveness of analytics.

3. Data Privacy and Security: With the increase in data breaches and cyber threats, ensuring the privacy
and security of sensitive business and customer data is paramount. Organizations must comply with
data protection regulations like GDPR and HIPAA, which can be challenging given the complexity and
volume of data they handle.

4. Skill Gap: There is often a significant gap between the demand for skilled business analytics
professionals and the available talent pool. Finding and retaining individuals with the necessary
analytics expertise and business acumen is a critical challenge for many companies.

5. Cost of Implementation: Deploying a robust business analytics infrastructure can be costly, especially
for small to mid-sized enterprises. The expense includes not just technology and tools, but also training
staff and maintaining the system.

6. Change Management: Integrating business analytics into the daily operations of an organization
requires significant change management. Employees need to adapt to new technologies and processes,
which can be met with resistance, necessitating effective communication and training strategies.

7. Technology Selection and Implementation: Choosing the right tools and technologies that fit the
specific needs of an organization is challenging. Additionally, implementing these technologies without
disrupting existing operations requires careful planning and execution.

8. Scaling Analytics Capabilities: As organizations grow, their data and analytics needs evolve. Scaling
analytics capabilities efficiently to handle increased data volume and complexity without performance
degradation is a significant challenge.

9. Generating Actionable Insights: Finally, the ultimate challenge is not just to perform analytics but to
derive actionable and relevant insights that can lead to effective decision-making. It requires not only
technical capabilities but also a deep understanding of business context and strategic objectives.

50 | P a g e
 Application of Business Analytics
Application of Business Analytics

Companies use Business Analytics (BA) to make data-driven decisions. The insight gained by BA enables these
companies to automate and optimize their business processes. In fact, data-driven companies that utilize
Business Analytics achieve a competitive advantage because they are able to use the insights to:

• Conduct data mining (explore data to find new patterns and relationships)

• Complete statistical analysis and quantitative analysis to explain why certain results occur

• Test previous decisions using A/B testing and multivariate testing

• Make use of predictive modeling and predictive analytics to forecast future results

Business Analytics also provides support for companies in the process of making proactive tactical decisions,
and BA makes it possible for those companies to automate decision making in order to support real-time
responses.

Challenges with Business Analytics

Penn State University’s John Jordan described the challenges with Business Analytics: there is “a greater
potential for privacy invasion, greater financial exposure in fast-moving markets, greater potential for mistaking
noise for true insight, and a greater risk of spending lots of money and time chasing poorly defined problems or
opportunities.” Other challenges with developing and implementing Business Analytics include…

• Executive Ownership: Business Analytics requires buy-in from senior leadership and a clear corporate
strategy for integrating predictive models

• IT Involvement: Technology infrastructure and tools must be able to handle the data and Business
Analytics processes

• Available Production Data vs. Cleansed Modeling Data: Watch for technology infrastructure that
restrict available data for historical modeling, and know the difference between historical data for model
development and real-time data in production

• Project Management Office (PMO): The correct project management structure must be in place in order
to implement predictive models and adopt an agile approach

• End user Involvement and Buy-In: End users should be involved in adopting Business Analytics and
have a stake in the predictive model

• Change Management: Organizations should be prepared for the changes that Business Analytics bring
to current business and technology operations

• Explainability vs. the “Perfect Lift”: Balance building precise statistical models with being able to
explain the model and how it will produce results

51 | P a g e