Industrial Engineering

(ME212)
Unit 4: FORECASTING

2
 INTRODUCTION
______________________________________________________________

Forecasting plays a crucial role in the development of plans

for the future. It is essential for the organisations to know
for what level of activities one is planning before
investments in inputs, i.e., men, machines and materials be
made.

Before making an investment decision, many questions will

arise like:
• What should be the size or amount of capital required?
• How large should be the size of the work force?
• What should be the size of the order and safety stock?
• What should be the capacity of the plant?

The answers to the above questions depends upon the

forecast for the future level of operations.
3
 INTRODUCTION
______________________________________________________________

Forecasting as defined by American Marketing Association is:

"An estimate of sales in physical units (or monetary value) for
·a specified future period under proposed marketing plan and
under the assumed set of economic and other forces outside
the organisation for which the forecast is made"

Forecasting is thus looked upon as a projection based on the

past data. Forecasting is not a guess work. It is the inference
based on large volume of data on past performance.

4
 FORECASTING AND PREDICTION
______________________________________________________________

Prediction is an estimate of future event through subjective

considerations other than the past data. For prediction, a
good subjective estimation is based on managers skill,
experience and judgement.
There is an influence of one's own perception and bias in
prediction. So it is less accurate and has low reliability.

Forecasting is based on the historical data and it requires

statistical and management science techniques. When we
refer to forecasting, usually it is combination of forecasting
and prediction.

5
 NEED FOR DEMAND FORECASTING
______________________________________________________________

1. Majority of the activities of the industries depend upon the

future sales.
2. Projected demand for the future assists in decision-making
with respect to investment in plant and machinery, market
planning and programmes.
3. To schedule the production activity and ensure optimum
utilisation of plant's capacity.
4. To prepare material planning to take up replenishment action
to make the materials available at right quantity and right time.
5. To provide an information about the relationship between
demand for different products in order to obtain a balanced
production in terms of quantity required of different products
as a function of time.
6. Forecasting is going to provide a future trend which is very
much essential for product design and development.
6
 LONG-TERM AND SHORT-TERM FORECASTS
______________________________________________________________

Depending upon the period for which the forecast is made, it is

classified as long-term forecasting and short-term forecasting.
Forecasts which cover the periods less than one year is termed
as short-term forecasting, and which cover the period over one
year (5 years or 10 years) future are termed as long-term
forecasts.

Short-term forecasts are made for the purpose of materials

control, loading and scheduling and budgeting. Long-term
forecasts are made for the purposes of product diversification,
sales and advertising budgets, capacity planning and investment
planning.

7
 CLASSIFICATION OF FORECASTING METHODS
______________________________________________________________
Qualitative
1. Judgemental techniques: Method which relies on the art of
human judgement. This is in practice since long time. The
other two techniques are relatively new and are heavily use
statistics for analysing the past data.
Quantative
1. Time Series methods: It identifies the historical pattern of
demand for the product and project or extrapolates this
demand into the future. The important method of making
inference about future on the basis of what has happened in
the past is called time series analysis.
2. Causal methods (Econometric Forecasting): The analyst tries
to establish cause and effect relationships between sales and
some other parameter that are related to sales, i.e., the
demand for cement depends upon the projected growth of
construction industry. It utilises regression and correlation
analysis. 8
 JUDGEMENTAL TECHNIQUES
______________________________________________________________

These are commonly used techniques in business and

industries. This is a subjective method where, there is a heavy
reliance on the past experience of the person and skill.

The various judgemental techniques are:

Customer and
Executive
Distributor Surveys
Opinion Method

Opinion Survey Qualitative

Market Research
Method Methods
9

Marketing Trials
Delphi Technique
9
 Executive Opinion Method
______________________________________________________________

• In this method, the expert's opinion is sought on the future

demand for the product.
• It is biased and subjective as it is not backed by any
scientific method or statistical data.
• The accuracy of the predicted demand depends upon the
skill, expertise and experience of the person making the
forecast.
• This method is used for demand forecasting of established
products.

10
 Opinion Survey Method:
______________________________________________________________

• Opinion method is relatively simple and practicable method

for forecasting demand for the new products.
• In this method, opinions are collected from the prospective
buyers regarding, why they buy a particular product, what
they expect from the product.
• The sampling technique is used to survey the customers.
• From the representative sample it is possible to forecast
how the targeted population responds to the product.

11
 Customer and Distributor Surveys
______________________________________________________________

• The individuals who have bought a product (customers) can

be asked the reasons for making the purchase.
• The questionnaire can be given along with the guarantee
cards.
• Estimates of expected sales (distributor surveys) can be
requested from retail outlets and company's sales force.
• Many companies heavily rely on judgements made by their
sales personnel.

12
 Marketing Trials
______________________________________________________________

• This method is specially applicable to new products.

• If the product is entirely new to the customer or market, then
it is very difficult to know the acceptability of the product
by the market.
• In this case it is advisable to expose the product to the
limited market trial.
• Such a trial is like a controlled experiment in which the
market area and the method of presentation are carefully
selected and controlled.
• Its cost is very high.
• This method is normally recommended for consumables
like cosmetics, tooth pastes, etc.

13
 Market Research
______________________________________________________________

• This method can be used for new products or existing

products.
• Usually the work is assigned to external marketing
agencies.
• This is recommended if extensive data is needed.
• The purpose of the research is to gather information
regarding the nature of consumption.
• The details about various factors which influence the
demand like location, buyer occupation, quantity,
quality, consumer income, etc., are collected and the
factors are related to get the forecast.

14
 Delphi Technique:
______________________________________________________________

• To make the judgemental forecasts more realistic by

minimizing bias, this method is used.
• In this method, a panel of experts are asked sequential
questions in which the response to one questionnaire is
used to produce next questionnaire.
• The information available to some experts are made
available to the other experts.
• This technique is an iterative process.

15
 Delphi Technique:
______________________________________________________________

16
 Delphi Technique:
______________________________________________________________

l. Choose the experts to participate representing a

variety of knowledgeable people in different areas
2. Through a questionnaire (or E-mail), obtain forecasts
from all participants
3. Summarize the results and redistribute them to the
participants along with appropriate new questions
4. Summarize again, refining forecasts and conditions,
and again develop new questions
5. Repeat Step 4 as necessary and distribute the final
results to all participants

17
 TIME SERIES ANALYSIS
______________________________________________________________

Time series refers to the past data arranged in a

chronological order with respect to time.

For example, the sales of TV sets for the last four years in a
particular geographical region are:

The time series methods does not study the factors that
influence the demand and in this method all the factors that
shape the demand are grouped into one factor-time and demand
is expressed as a series of data with respect to time.
18
 TIME SERIES ANALYSIS
______________________________________________________________

The time series analysis consists of determining the trend

underlying the demand and extrapolate the future trend.
Statistical methods are used to determine the trend. Time series
consists of four components:

1. Trend (T): The long period tendency of the data to

increase or decrease is referred to as a secular trend. It is
the long run historical component of the time series which
indicates overall growth or decline of the business
overtime.
2. Cyclical Fluctuations (C): are the oscillating patterns of
increase or decrease around the trend. Cyclical fluctuations
account for some of the variance between the trend line and
the data points. The magnitude, timing and pattern of cyclic
fluctuations vary so widely and the awareness of the effect of
cyclic variation should influence the forecaster. 19
 TIME SERIES ANALYSIS
______________________________________________________________

3. Seasonal Variations (S): This seasonal variation occurs with

some degree of regularity in a span of one year and are
annually repetitive. These variation are caused by climatic
conditions, social customs, festivals, e.g., the sales of paint is
highest during Diwali Festival, sale of umbrella and rain coat
during monsoon, and warm clothes during winter.

4. Irregular Variations (R): These occur without any fixed

pattern. These are chance variations that are not explained by
trend, cyclic or seasonal variations. These may account for
natural calamities like flood, draught, earthquake, etc.

20
 Seasonal variation

x
x x Linear
x x
x x Trend
x
Sales

x
x x x
x
x
xx
x xx x x
x
x
x x x x x x
x x x x x x
x x x
x xxxxx
x
x x

1 2 3 4
Year

21
SIMPLE MOVING AVERAGE (SMA) FORECASTING
______________________________________________________________

• The past data of sales of a company can have fluctuations

(high or low) because of the seasonal variations and random
variations.
• Simple averaging of demand for previous periods is going to
hide the trend and it is meaningless since trend is an
important factor.
• Simple Moving Average (SMA) consists of series of
arithmetic means calculated from overlapping groups of
successive elements of time series.
• This method uses a past data and calculates a rolling
average for a constant period.
• At each period, fresh average is computed at the end of
each period by adding the demand of the most recent
period and deleting the data of the old-period since the
data in this method changes from period to period, it is
called moving average method. 22
SIMPLE MOVING AVERAGE (SMA) FORECASTING
______________________________________________________________

 The simple moving average model assumes an average

is a good estimator of future behavior
 The formula for the simple moving average is:

A t-1 + A t-2 + A t-3 +...+A t- n

Ft =
n

Ft = Forecast for the coming period

N = Number of periods to be averaged
A t-1 = Actual occurrence in the past period for up to “n”
periods

23
 SIMPLE MOVING AVERAGE (SMA) FORECASTING
______________________________________________________________

A t-1 + A t-2 + A t-3 +...+A t- n

Ft =
n
Week Demand
Question: What are the 3-week
1 650
and 6-week moving average 2 678
forecasts for demand? 3 720
Assume you only have 3 weeks 4 785
and 6 weeks of actual demand 5 859
6 920
data for the respective 7 850
forecasts 8 758
9 892
10 920
11 789
12 844
24
SIMPLE MOVING AVERAGE (SMA) FORECASTING
______________________________________________________________
Calculating the moving averages gives us:
Week Demand 3-Week 6-Week
1 650 F4=(650+678+720)/3
2 678 =682.67
3 720 F7=(650+678+720
4 785 682.67 +785+859+920)/6
5 859 727.67 =768.67
6 920 788.00
7 850 854.67 768.67
8 758 876.33 802.00
9 892 842.67 815.33
10 920 833.33 844.00
11 789 856.67 866.50
12 844 867.00 854.83 25
 SIMPLE MOVING AVERAGE (SMA) FORECASTING
______________________________________________________________

Plotting the moving averages and comparing them shows

how the lines smooth out to reveal the overall upward trend
in this example

Note how the

950
900 3-Week is
850
d 800 Demand
smoother than
n 750
a
m 700 3-Week the Demand,
e
D 650 6-Week and 6-Week is
600
550 even smoother
500
1 2 3 4 5 6 7 8 9 10 11 12

Week
Period n of moving average should
. be carefully selected.
The wrongly selected period will distort the data and gives
wrong picture of the trend. Larger the period of moving
average, the greater is the smoothing effect 26
 Weighted MOVING AVERAGE FORECASTING
________________________________________________________

In simple MA, equal weightage is given to 1st month, 2nd month

and 3rd month in a three month moving average. But the
organization wants to attach more weightage to the third month
and least to the first month.
The formula for the moving average is:

Ft = w 1 A t -1 + w 2 A t - 2 + w 3 A t -3 + ...+ w n A t - n

wt = weight given to time period “t” n

occurrence (weights must add to one) w

i=1
i =1
.

27
Weighted MOVING AVERAGE FORECASTING
________________________________________________________

Question: Given the weekly demand and weights, what is

the forecast for the 4th period or Week 4?

Week Demand Weights:

1 650
2 678
t-1 .5
3 720 t-2 .3
4 t-3 .2

Note that the weights place more emphasis on the

most recent data, that is time period “t-1”
.

28
Weighted MOVING AVERAGE FORECASTING
________________________________________________________

Week Demand Forecast

1 650
2 678
3 720
4 693.4

Ft = w 1 A t-1 + w 2 A t-2 + w 3 A t-3 +...+w n A t- n

.
F4 = 0.5(720)+0.3(678)+0.2(650)=693.4

29
Weighted MOVING AVERAGE FORECASTING
________________________________________________________

Question: Given the weekly demand information and

weights, what is the weighted moving average forecast
of the 5th period or week?

Week Demand Weights:

1 820 t-1 .7
2 775 t-2 .2
3 680
t-3 .1
4 655

30
Weighted MOVING AVERAGE FORECASTING
________________________________________________________

Problem: The past data on the load on the weaving machines,

is shown. below:

(a) Compute the load on the weaving machine centre using 5th
moving average for the month of. December 1996.
(b) Compute a weighted three months moving average for December,
1996 where the weights are 0.5 for the latest month, 0.3 and 0.2 for
the other months respectively. 31
Weighted MOVING AVERAGE FORECASTING
________________________________________________________

Problem: The past data on the load on the weaving machines,

is shown. below:

(a) Compute the load on the weaving machine centre using 5th
moving average for the month of. December 1996. → 773hrs
(b) Compute a weighted three months moving average for December,
1996 where the weights are 0.5 for the latest month, 0.3 and 0.2 for
the other months respectively. → 947.8 hrs 32
Weighted MOVING AVERAGE FORECASTING
________________________________________________________

Problem: The data given below represents sales figures of ABC

company for the 12 months of the year 1996.
1. Compute 3 months moving average (ignoring decimal values)
2. Forecast the demand for the month of Jan. 1997
3. If the actual demand for the month of Jan. 1997 is 905 units.
what should be the forecast for the month of Feb. 97.

33
 Exponential SMOOTHING FORECASTING
________________________________________________________

• One of the disadvantages of the moving average forecasting is

the laborious operation of maintaining the data for all the
previous years.

• Exponential smoothing method requires only the current

demand and the forecasted demand for the current month.

• Demand for the most recent data is given more weightage

and the weights assigned to older periods decrease
exponentially.

• Thus exponential forecasting ensure that the forecast made

by this method keeps pace .with changing business trend.

34
 Exponential SMOOTHING FORECASTING
________________________________________________________

Simple exponential smoothing model estimates the average

forecast for the next period by using the actual and the
forecasted demand for the previous period.

The basic advantage of this method over the moving average

method is that, one needs only two figures - one for the old
forecast and another for the actual sales observation.
. determined by:
If the value of α is not given then it can be

35
 Exponential SMOOTHING FORECASTING
________________________________________________________

Week Demand Question: Given the

1 820 weekly demand
2 775 data, what are the
3 680 exponential
4 655
smoothing
5 750
forecasts for
6 802
7 798
periods 2-10 using
8 689 a=0.10 and
9 775 a=0.60?
10 Assume F1=D1
.

36
 Exponential SMOOTHING FORECASTING
________________________________________________________
Answer: The respective alphas columns denote the forecast values. Note
that you can only forecast one time period into the future.

Week Demand 0.1 0.6

1 820 820.00 820.00
2 775 820.00 820.00
3 680 815.50 793.00
4 655 801.95 725.20
5 750 787.26 683.08
6 802 783.53 723.23
7 798 785.38 770.49
8 689. 786.64 787.00
9 775 776.88 728.20
10 776.69 756.28 37
 Exponential SMOOTHING FORECASTING
________________________________________________________

Note how that the smaller alpha results in a smoother line in

this example

850
800
750 Demand
Demand

700
0.1
650
600 0.6
550
500
1 2 3 4 5 6 7 8 9 10
Week
.

The value of the α selected is small (0.05 to 0.1), if the demand pattern is smooth or stable
and large value of α is used for the fluctuating demand. 38
 Exponential SMOOTHING FORECASTING
________________________________________________________

39
 Exponential SMOOTHING FORECASTING
________________________________________________________

Problem: The demand for the disposable plastic tubing for a

general hospital is 300 units and 350 units for September and
October respectively. Using 200 units as forecast for
September, compute the forecast for the month of November.
Assume the value of a as 0.7.

40
 Exponential SMOOTHING FORECASTING
________________________________________________________

Problem: Estimate the sales forecast for the year 2000,

using exponential smoothing forecastor. Take α = 0.5 and
the forecast for the year 1995 as 160 x 105 units. Compare
the forecast with least square method.

41
 CAUSAL FORECASTING METHOD
________________________________________________________

• Casual methods try to identify the factors which causes

the variation of demand and try to establish a
relationship between the demand and these factors.

• In the method, the analyst tries to identify those factors

that best explain the level of sales of the product. This
process is called econometric forecasting.

• The objective of this method is to establish a cause

and effect relationship between the changes in the
.
sales level of the product and set of relevant
explanatory variable.

42
CAUSAL FORECASTING: Regression and Correlation
________________________________________________________

Regression: Regression analysis is a forecasting technique

that establishes the relationship between variables.
Historical data establishes a functional relationship between
the two variables.

Correlation:Correlation analysis is determining the degree

of closeness of the relationship between two variables.

In general, the sales of any product is influenced by

external or internal factors. Sales is thus a dependent
variable and is a function of one or more independent
variable. A regression models (simple or multiple)
.
establish the relationship between the independent and
dependent variable.

43
 CAUSAL FORECASTING: Regression
________________________________________________________
Linear Regression: This is the mathematical method of
obtaining the line of best fit between the dependent variable
(usually demand) and an independent variable (usually
time). This method is called least square method as the sum
of the square of the deviations of the various points from the
line of best fit is minimum or least.
In a simple regression analysis, the relationship between the
dependent variable y and some independent variable x can be
represented by a straight line.

44
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

45
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

46
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________
Problem: The following data gives the sales of the company
for various years. Fit the straight line. Forecast the sales for
the year 1998 and 1999.

47
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

48
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

49
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

Problem: The sales for the domestic water pumps manufactured by

Ajit Manufacturing Company is given forecast the demand for the
pumps for the next three years using least square method.

50
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

51
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

52
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

53
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

a= 15.07 b = 3.11 y =. 155.02

54
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________
Arnold Tofu owns and operates a chain of 12 vegetable protein "hamburger"
restaurants in northern Louisiana. Sales figures and profits for the stores are in the table
below. Sales are given in millions of dollars; profits are in hundreds of thousands of
dollars. Calculate a regression line for the data. What is your forecast of profit for a
store with sales of $24 million? $30 million?

$4,004,000
. $4,856,600

55
 LEAST SQUARE METHOD OF
FORECASTING (Regression Analysis)
________________________________________________________

Problem • Estimate the sales forecast for the year 2000,

using exponential smoothing forecastor. Take alpha = 0.5
and the forecast for the year 1995 as 160 x /05 units.
Compare the forecast with least square method.

56
 Correlation FORECASTING METHOD
________________________________________________________

• Correlation analysis is determining the degree of

closeness of the relationship between two variables.

•Co-efficient of correlation
The degree of relationship is called correlation. It is a single figure
which expresses the degree and direction of correlation is called co-
efficient of correlation.

57
 Correlation FORECASTING METHOD
________________________________________________________

58
 Correlation FORECASTING METHOD
________________________________________________________

Coefficient of Determination (γ2)

The coefficient of determination is the percent of variation in the

dependent variable (y) that is explained by the regression equation.

If value is 0.81, indicating that 81% of the total variation is explained by

the regression equation.

59
 Correlation FORECASTING METHOD
________________________________________________________

60
 Correlation FORECASTING METHOD
________________________________________________________

61
Nodel Construction Company renovates old homes in West
Bloomfield, Michigan. Over time, the company has found that its
dollar volume of renovation work is dependent on the West
Bloomfield area payroll.

a.) Management wants to establish a mathematical relationship to help predict sales.

b) The VP now wants to know the strength of the association between area payroll and
sales.

62
Monitoring and Controlling Forecasts

63
 FORECAST ERROR
________________________________________________________

The demand for the product is forecasted using many

forecasting methods. It is essential to have a good
measure of effectiveness of the methods.

Forecast error is the numerical difference between the

forecasted demand and the actual demand. The error
should be minimum as far as possible.

64
 Mean Absolute Deviation (MAD)
________________________________________________________
It is a measure of forecast error and it measures the average
forecast error without direction.
It calculate all the sum of the absolute value of the
forecast error for all periods divided by the total number
of periods.

In MAD, errors are measured without considering signs.

Thus it expresses the magnitude
. but not the direction of error.

This absolute value is referred as Mean Absolute Deviation.

65
BIAS/ Average Error Estimate/ Mean Forecast error
________________________________________________________
This measures the forecast error with regard to direction and
shows any tendency to over forecast or under forecast.
This is also calculated as sum of forecast error for all periods
divided by the total number of periods.

BIAS indicates the directional. tendency of forecast errors.

If the forecast repeatedly overestimates actual demand, BIAS
will have a positive value and underestimation will be
indicated by a negative bias. 66
________________________________________________________

Problem 11: A dealer for electrical appliances forecasts the

demand for the Geyser at the rate of 500 per month for the
next three months. The actual demands turned out to be
400,560 and 700.

Calculate the MAD and bias, comment on the same.

67
 Mean Square error (MSE)
________________________________________________________

It is the mean of the squares of the deviations of forecast

demands from the actual demand values.

68
 Running Sum forecast error (RSFE)
________________________________________________________

Cumulative deviations error/ cummalative sum error or

Running Sum of Forecast Error (RSFE)

This is the algebraic sum of the forecasting error i.e. both

positive and negative signs are considered.

69
 Mean Absolute Percentage Error (MAPE)
________________________________________________________

Mean absolute percentage error is the mean of the

percentage deviation of the forecast demands from the
actual demands.

70
 Tracking signal
________________________________________________________

The tracking signal tells how well the forecast is predicting

actual values.

Positive tracking signals indicate that demand is greater than

forecast. Negative signals mean that demand is less than forecast.
A good tracking signal—that is, one with a low cumulative
error—has about as much positive error as it has negative error.

71
Carlson’s Bakery wants to evaluate performance of its
croissant forecast.
Develop a tracking signal for the forecast, and see if it stays
within acceptable limits, which we define as ± 4 MADs.

72
73
 Problem
________________________________________________________

The Forecast and the actual demand for the 6 months is

given below. Compute the various Forecast errors and
tracking signal and comment on the forecast accuracy.

74
Problem 13: Starwars Co. Ltd., uses simple exponential smoothing
with smoothing constant α = 0.2 to forecast the demand. The
forecast for the first week of March was 400 units and the actual
demand turns out to be 450 units.
(i) Estimate the demand for the second week of March.
(ii) If the actual demand for the second week of March is 460 units,
forecast the demand upto April second week. Assume that demand
for subsequent weeks are 465, 434, 420, 498, 462 and 470 unit.

75
During the past 8 quarters, the Port of Baltimore has unloaded large quantities of grain
from ships. The port’s operations manager wants to test the use of exponential
smoothing to see how well the technique works in predicting tonnage unloaded. He
guesses that the forecast of grain unloaded in the first quarter was 175 tons. Two values
of α are to be examined: α = .10 and α = .50.

Evaluate the accuracy of each smoothing constant using MAD, MSE, MAPE.

76
 Exponential Smoothing with Trend Adjustment
Here is why exponential smoothing must be modified when a trend is present. Assume
that demand for our product or service has been increasing by 100 units per month and
that we have been forecasting with a = 0.4 in our exponential smoothing model.

To improve our forecast, let us illustrate a more complex exponential smoothing

model, one that adjusts for trend.

Forecast including trend (FITt) = Exponentially smoothed

forecast average (Ft) + Exponentially smoothed trend (Tt)

77
With trend-adjusted exponential smoothing, estimates for both the
average and the trend are smoothed. This procedure requires two
smoothing constants: α for the average and β for the trend.
Ft = α(Actual demand last period) + (1 - α)(Forecast last period + Trend
estimate last period)
Ft = α(Dt-1) + (1 - α)(Ft-1+ Tt-1)
Tt = β(Forecast this period - Forecast last period) + (1 - β)(Trend estimate
last period)
Tt = β(Ft - Ft-1) + (1 - β) Tt-1
where Ft = exponentially smoothed forecast average of the data series in period t
Tt = exponentially smoothed trend in period t
Dt = actual demand in period t
α = smoothing constant for the average (0 ≤ α ≤ 1)
β = smoothing constant for the trend (0 ≤ β ≤ 1)

78
Dr. Susan Sweeney, a providence psychologist, specializes in treating
patients who are agoraphobic (i.e. afraid to leave their homes). The
following table indicates how many patients. Dr. Sweeney has seen each
year for the past 10 years. It also indicates what the robbery rate was in
Providence during the same
year:

(i) Plot these data and decide if a linear model is reasonable.

(ii) Develop a regression relationship.
(iii) What is expected number of patients Dr. Sweeney will see in years 11 and 12.
(iv) What is standard error of the estimate?
(v) What is the model’s correlation coefficient and coefficient of determination?

82
Dell uses the CR 5 chip in some of its laptop computers. The prices for the
chip during the last 12 months were as follows:

(i) Use a 2-month moving average on all the data and plot the averages and
the prices.
(ii) Use a 3-month moving average and add the 3-month plot to the graph
created in part (i).
(iii) Which is better (using the mean absolute deviation): the 2-month average
or the 3-month average?

83
 Seasonal Variations in Data
1. Find the average historical demand each season (or month in this case) by
summing the demand for that month in each year and dividing by the number of
years of data available. For example, if, in January, we have seen sales of 8, 6, and
10 over the past 3 years, average January demand equals (8 + 6 + 10)/3 = 8
units.
2. Compute the average demand over all months by dividing the total average
annual demand by the number of seasons. For example, if the total average demand
for a year is 120 units and there are 12 seasons (each month), the average monthly
demand is 120/12 = 10 units.
3. Compute a seasonal index for each season by dividing that month’s historical
average demand (from Step 1) by the average demand over all months (from
Step 2). For example, if the average historical January demand over the past 3
years is 8 units and the average demand over all months is 10 units, the seasonal
index for January is 8/10 = .80. Likewise, a seasonal index of 1.20 for February
would mean that February’s demand is 20% larger than the average demand over
all months.
4. Estimate next year’s total annual demand.
5. Divide this estimate of total annual demand by the number of seasons, then
multiply it by the seasonal index for each month. This provides the seasonal
forecast .
84
A Des Moines distributor of Sony laptop computers wants to develop monthly indices for
sales. Data from the past 3 years, by month, are available.

85
86
If we expect the annual demand for computers to be 1,200 units next year, we would
use these seasonal indices to forecast the monthly demand as follows:

87