Scribd
Upload
256 views19 pages

1-Introduction To Business Forecasting

This document provides an overview of business forecasting techniques. It discusses that business decisions depend on forecasts, and the need for forecasting is growing, with many companies now having dedicated forecasting departments. Forecasting is used not just for private businesses but also in other sectors to predict things like migration, crime rates, housing starts, etc. While quantitative forecasting can be complex, computer packages now make the process easier. The key steps in forecasting are specifying objectives, determining what to forecast, identifying the time dimension, collecting and preparing data, selecting a model, evaluating the model, preparing forecasts, and tracking results. Time series data used in forecasting often show trends, seasonal patterns, cyclical patterns, and irregular fluctuations. Different

Uploaded by

Stephen Le
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
256 views19 pages

1-Introduction To Business Forecasting

This document provides an overview of business forecasting techniques. It discusses that business decisions depend on forecasts, and the need for forecasting is growing, with many companies now having dedicated forecasting departments. Forecasting is used not just for private businesses but also in other sectors to predict things like migration, crime rates, housing starts, etc. While quantitative forecasting can be complex, computer packages now make the process easier. The key steps in forecasting are specifying objectives, determining what to forecast, identifying the time dimension, collecting and preparing data, selecting a model, evaluating the model, preparing forecasts, and tracking results. Time series data used in forecasting often show trends, seasonal patterns, cyclical patterns, and irregular fluctuations. Different

Uploaded by

Stephen Le
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 19

Overview of business

forecasting
Chapters 1&2

Business Forecasting
Business decisions always depends on
some forecast about the course of an
event. The sales forecast is the root
forecast from which other forecasts like
employment, etc. forecast is derived.
The need for personnel with forecasting is
growing . Many companies now have a
forecasting departments with large group
of trained staff.
2

Forecasting
The forecasting need is not limited to
private companies and forecast of
sales, in other sections of economy
and science people make decisions
based on forecasts like forecast of
migrations to a state, forecast of
homicides, drug use, housing starts,
etc.
3

Forecasting
Although quantitative forecasting is a
complicated job, the use of computer
packages makes it possible and easy.

The Forecast Process

1. specifying objectives
2. Determining what to forecast.
3. Identifying time dimension.
4. Data consideration
5. Model selection
6. Model evaluation
7. Forecast preparation
8.Tracking the results.
5

Trend, Seasonal, and Cyclical


Pattern
The data that are used most often in
forecasting is time series.
Time series data can show a variety
of patterns:
Trend
Seasonal
Cyclical
Irregular
6

Data pattern and Model


selection
Smoothing models
Best used for the short-term forecasts
when there is seasonality or trend.

Trend regressions.
Causal Regression Model
Best used for models with cyclical
movements, trends.

Time-series Models
7

A Statistical Review
We need to forecast when future
outcome can happen with different
probabilities.
Dealing with probabilities, the first
thing that we are interested to know is
the what is the central tendency of the
distribution.
The Three most measures of the central
tendency are mean, median, mode.
8

Mean, Median, Mode


The mean is the arithmetic average.
The median is the value that splits
the responses into two equal parts
when they are arrayed form the
smallest to largest.
The mode is the response that occurs
most frequently.

Normal distribution
In addition to information about
central tendency, we would like to
know how the observations are
distributed around the mean.
For the purpose of forecasting we
usually concentrate on two
distributions: normal and T
distribution.
10

Normal Distribution
Is symmetric around the mean. It has the
following important properties,
The area between one standard deviation from
the mean is 68%.
The area between two standard deviations
from the mean is 95%.
The area between three standard deviation
from the mean is 99%.

11

Standard Normal
Distribution
Suppose we know Mean= = 50, = 10,
we want to know what is the probability
that sales is between 35 and 60.

X
Z

Z= (35-50)/10 = -1.5
Z= (60-50)/10 =
1
Probability =0.4332+0.3413= 0.7745
12

Sample mean
Suppose we take a sample and its mean is
250.
We would like to test if the population mean
can be 280.
First we change the value
to standard
X
Z
/ n
normal

N=100, = 50,

=250
13

Hypothesis testing

The sample mean for number of homicide in a small city is


250 in year. The population standard deviation is 100. Test
the hypothesis that the population mean is 280 at 5% level
of significance.

H0: Pop mean = 280

H1: pop mean # 280


Z = (250-280)/(50/10)= -6
Rejection area 1.96, -1.96
We reject the null that the
mean is 250

14

One side test against two side


tests
Two sided tests
Null: X= a
Alter: X a
The significance level will be divided by 2

One sided test


Null: X>= a
Alter: X< a
The significance level will be on one side
15

Type one and two error


Type one error = possibility of
rejecting the null when null is correct
Type two error: possibility of
accepting the null when null is
wrong.

16

T distribution
When population standard deviation is
unknown we use t distribution.

X
s/ n

17

Example
ACC corporation produces a popular PC clone .
The sales manager for ACC has recently read a
report that indicated that sales per salesperson
for other producers are normally distributed with
a mean of $255000. she is interested in knowing
whether her sales stall is comparable. She picked
a random sample of 16 salespeople and obtained
the following results
Microsoft Excel
Worksheet

18

Type one and type two error


What is the type one error in the
previous example?
What is the type two error in the
previous example?
Which one should be minimized?

19

You might also like