Scribd
Upload
7 views

Session 09 Lecture Notes 0215

The document provides an overview of key statistics topics, including descriptive and inferential statistics, estimation, sampling distributions, and the Central Limit Theorem (CLT). It explains the importance of sampling methods, population parameters, and how to make statistical inferences based on sample data. Additionally, it includes an example illustrating the application of the CLT in calculating probabilities related to sample means.

Uploaded by

aneesenjpg518
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
7 views

Session 09 Lecture Notes 0215

The document provides an overview of key statistics topics, including descriptive and inferential statistics, estimation, sampling distributions, and the Central Limit Theorem (CLT). It explains the importance of sampling methods, population parameters, and how to make statistical inferences based on sample data. Additionally, it includes an example illustrating the application of the CLT in calculating probabilities related to sample means.

Uploaded by

aneesenjpg518
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

Links relevant for the session

▪ Statistics overview - Link


▪ Population and Sample - Link
▪ Point Estimation - Link
▪ Sampling Distribution - Link
▪ CLT Real World Application – Link
▪ Mathematical proofs for CLT outcomes - Link
Topics Covered
▪ Statistics Topics Overview
▪ Estimation
▪ Sampling Distributions & Central Limit Theorem
▪ Confidence Level and Confidence Interval
Core Topics in Statistics

Descriptive Statistics Inferential Statistics


✓ Measure of Central Tendency
▪ Mean, median, Mode
✓ Measure of Dispersion or Variation
▪ Variance, Standard Deviation
▪ Range, Percentile
▪ Skewness & Kurtosis
✓ Measure of Distribution
▪ Count/ Frequency Distribution
▪ % Frequency Distribution
Core Topics in Statistics

Descriptive Statistics Inferential Statistics


✓ Measure of Central Tendency ✓ Population and Samples
▪ Sample selection with randomization
▪ Mean, median, Mode
✓ Making estimation about Population using
✓ Measure of Dispersion or Variation samples
▪ Variance, Standard Deviation ▪ Point estimates
▪ Range, Percentile ▪ Interval estimates (confidence intervals)
▪ Errors associated with these estimations
▪ Skewness & Kurtosis
✓ Concept of Hypothesis testing
✓ Measure of Distribution ▪ Null and Alternate Hypothesis
▪ Count/ Frequency Distribution ▪ Statistical significance, p-value and Errors
▪ % Frequency Distribution ✓ Different Statistical Tests
▪ z –test, t – test , f - test
▪ Chi-square and ANOVA tests
Estimation
Population and Samples

By working on the drawn samples, Inferences are made about the entire population

Population Sample

Population characteristics are known as parameters Sample characteristics are called statistics
mean 𝜇 , standard deviation (𝜎) mean 𝑥ҧ , standard deviation (𝑠)

▪ A population is a collection of ‘ALL’ objects, ▪ These samples are created using different
people or data points for which we want to sampling methods
draw conclusion about ✓ Random Sampling
▪ A sample on the other hand is a finite ✓ Stratified Sampling
collection (of objects, people or data points) ▪ Various tests are conducted to ensure that the
considered from the population sample comes from the same population
Precursor to Sampling Distribution and Estimation

Sample D

Sample B Can we use one of these samples to


estimate population parameters ?

Sample A How close or far will the sample statistics


Sample C be to the actual population parameters?

What is the associated error to each of


these sample statistics ?
Sampling Distribution
and Central Limit Theorem
Sampling Distribution
Repeated samples (of size n each) are drawn. Means of these samples give the below Sampling Distributions (for n =2, 5 and 30)

Sample Size = 2

Sampling Distributions of Sample Means (𝑥)ҧ


Population

Sample Size = 5

Sample Size = 30

Similar to Sampling distribution of sample Means, we can have Sampling distribution of sample Medians, Standard Deviations or Variances
Intuition behind – Why Std. Deviation decreases with increase in sample size
Sampling Distribution
Repeated samples (of size n each) are drawn. Means of these samples give the below Sampling Distributions (for n =2, 5 and 30)

▪ Parameter estimates of a Population based on


Sample Size = 2 statistics (𝑥,ҧ 𝑠) of a single sample alone can be

Sampling Distributions of Sample Means (𝑥)ҧ


heavily affected by sampling error
▪ Using the mean statistics of multiple samples, a
Population distribution is created which is the ‘sampling
distribution of sample means’
Sample Size = 5 ▪ Mean (ഥ𝒙) of this distribution (sampling distribution
of sample means) is very close to the actual mean
of the population

Sample Size = 30

Similar to Sampling distribution of sample Means, we can have Sampling distribution of sample Medians, Standard Deviations or Variances
Central Limit Theorem

▪ For a sample size which is sufficiently larger, the sampling n=2


distribution of sample means will be approximately Normal
n=5
(irrespective of the distribution of population) - This is the
n=30
central limit theorem

▪ Here, the sampling distribution with larger sample size will


show a more compact normal distribution (imagine sampling
distribution for a sample of size = Population, which will be a single
point)

𝜎
ന=𝜇
𝑿 𝝈𝑿ഥ =
𝑛
Mean of the sampling Std. Dev of the
distribution sampling distribution
Why is Central Limit Theorem Important ?

We can draw Statistical Inferences that are based on a Normal


Distribution, even if we are sampling from a non-Normal
Distribution – provided we have a large size
Application of CLT : Example Question 01
Example : The distribution of Annual Earnings of all Bank Tellers with Five years of experience is shown below. This distribution
has a mean of $19,000 and a standard deviation of $2,000. If we draw a random sample of 25 tellers, what is the probability that
their earnings will average more than $19,750 annually.

𝜎 = $2,000

𝜇 = $19,000
Incorrect Approach to solve this problem
Application of CLT : Example Question 01
Example : The distribution of Annual Earnings of all Bank Tellers with Five years of experience is shown below. This distribution
has a mean of $19,000 and a standard deviation of $2,000. If we draw a random sample of 25 tellers, what is the probability that
their earnings will average more than $19,750 annually.

𝜎 = $2,000

𝜇 = $19,000
Application of CLT : Example Question 01
Example : The distribution of Annual Earnings of all Bank Tellers with Five years of experience is shown below. This distribution
has a mean of $19,000 and a standard deviation of $2,000. If we draw a random sample of 25 tellers, what is the probability that
their earnings will average more than $19,750 annually.

Sampling Distribution of Sample Means

$2,000
𝝈𝒙ഥ =
𝜎 = $2,000 25

𝜇 = $19,000 𝜇𝑥ҧ = $19,000

𝑥ҧ − 𝜇𝑥ҧ
𝑍=
𝜎𝑥ҧ
19,750 − 19,000
𝑍= 𝑃(𝑍 > 1.88)
400
𝑍 = 1.88
0 𝑍 = 1.88

You might also like