1

A Simple Moving Average Example

Moving Average Forecast =

demand in previous n periods

n

Where n is the number of periods in the moving average

Example: The demand for Gingerbread Men is shown in the table. Forecast the demand for
month 7.
Month
1
2
3
4

Demand
1500
2200
2700
4200

7800

5400

Calculation

Forecast

1500 2200 2700

2133
3

2133

Bakery Gingerbread Man Sales

9000
8000
7000

Sales

6000
5000

Demand

4000

Forecast

3000
2000
1000
0
1

4
Period

A Weighted Moving Average Example

Weighted Moving Average Forecast =

(weight for period n) (demand in period n)

weights

Example: The demand for Gingerbread Men is shown in the table. Forecast the demand for
month 7, weighting the past three months as follows: last month 3, two months ago 2, three
months ago 3
Period
Last Month
Two Months Ago
Three Months Ago
Sum of Weights
Month
1
2
3
4

Demand
1500
2200
2700
4200

7800

5400

Weight
3
2
1

Calculation

Forecast

(1500 1) (2200 2) (2700 3)

2333
6

2333

3
Bakery Gingrbread Man Sales
9000

8000

7000

6000

5000

Series1
Series2

4000

3000

2000

1000

0
1

P eriod

An Exponential Smoothing Example

Exponential Smoothing Forecast Ft = Ft-1 + (At-1 Ft-1)
Where

Ft
Ft-1

At-1

= New Forecast
= Previous Forecast
= Smoothing Constant: (0 1)
= Previous Periods Actual Demand

Example: The demand for Gingerbread Men is shown in the table. Forecast the demand for
month 7, using a smoothing constant of 0.4. The forecast for month 1 is 1500 units.
Month
1
2

Demand
1500
2200

Calculation
1500 + 0.4(1500 1500)

Forecast
1500
1500

2700

1500 + 0.4(2200 1500)

1780

4200

7800

5400

4
Bakery Gingerbread Man Sales
9000
8000
7000

Sales

6000
5000

Demand
Forecast

4000
3000
2000
1000
0
1

Period

A Mean Absolute Deviation Example

MAD =

forecast errors
n

Example: Calculate the MAD for the value of used in the Exponential Smoothing Example.
Month
1
2
3

Demand
1500
2200
2700

Forecast
1500
1500
1780

4200

2148

7800

2969

5400

4901

Error
1500 1500 = 0
2200 1500 = 700

| Error |
0
700

Total:
MAD:

A Mean Squared Error Example

5
MSE =

forecast errors

Example: Calculate the MAD for the value of used in the Exponential Smoothing Example.
Month
1
2
3

Demand
1500
2200
2700

Forecast
1500
1500
1780

4200

2148

7800

2969

5400

4901

Error2
0
490000

Error
1500 1500 = 0
2200 1500 = 700

Total:
MSE:

An Exponential Smoothing With Trend Adjustment Example

Forecast Including Trend Ft = (At-1) + (1 - )(Ft-1 + Tt-1)
Trend Tt = (Ft Ft-1) + (1 - )T t-1
Where

Ft
Tt
At

= Exponentially smoothed forecast for period t

= Exponentially smoothed trend for period t
= Actual demand for period t
= Smoothing constant for the average (0 1)
= Smoothing constant for the trend (0 1)

Example: The demand for Gingerbread Men is shown in the table. Forecast the demand for
month 7, using a smoothing constant for the average of 0.4, and a smoothing constant for the
trend of 0.2. The forecast for month 1 is 1500 units and the trend for month 1 is 200 units.
Month
1
2
3

Demand Forecast
1500
1500
2200
0.4(1500) + 0.6(1700)
= 1620
2700

4200

7800

Trend
200
0.2(1620 1500) + 0.8(200)
= 184

FIT
1700
1804

6
6

5400

B akery Gingerbread M an Sales

9000

8000

7000

6000
Demand

5000

Forecast
Trend

4000

FIT

3000

2000

1000

0
1

P er i od

Least Squares Trend Projection Example

a bx
y

Where

a
b
x

= computed value of variable to be predicted (ie dependant variable)

= y-axis intercept
= slope of regression line
= independent variable

We can determine a and b with the equations:

b

xy - nxy
x nx
2

a y - bx

Example: The demand for Gingerbread Men is shown in the table. Forecast demand for
period 7 by fitting a single line trend to the data.
Month (x)
1
2
3

Demand (y)
1500
2200
2700

4200

7800

x2
1
4

xy
(1)(1500) = 1500
(2)(2200) = 4400

7
6

5400

x =

y =

x2 =

xy - nxy
x nx
2

xy =

y
n

a y - bx =

Thus, our trend equation is:

To calculate the forecast for month x = 7, we have:

Seasonal Forecast Example

Average Monthly Demand =

Average Annual Demand

12 months

Average Annual Demand

Seasonal Index = Average Monthly Demand

Example: The demand for gingerbread men over the past three years is shown in the table. If
we expect the total yearly demand in 2002 to be 45,000 units, what will be our forecasted
monthly demands in 2002?
Month 1999
1
1100
2
1800

2000
1300
2000

2001
1500
2200

2300

2500

2700

3800

4000

4200

4500

4700

4900

5000

5200

5400

5500

5700

5900

4800

5000

5200

Average Annual Demand

(1100 + 1300 + 1500) / 3 = 1300

8
9

3000

3200

3400

10

2200

2400

2600

11

1500

1700

1900

12

1200

1400

1600
Average Monthly Demand =

Seasonal Index for January = 1300 /

Forecast for January 2002 =

45000

12

Regression Analysis Example

b

a bx
y

xy - nxy
x nx
2

a y - bx

Example: We think that there may be a relationship between park attendance and number of
gingerbread men sold. Data for the first six months are shown in the table. Forecast the
number of gingerbread men that will be sold in month 7 if monthly park attendance is forecast
as 25000 people.
Sales (y)

x2

xy

1
2

Attendance (x)
(,000)
8
12

1500
2200

64

(8)(1500) = 12000

14

2700

18

4200

19

7800

22

5400

x =

y =

x2 =

xy =

Month

9
x

xy - nxy
x nx
2

y
n

a y - bx =

Thus, our regression equation is:

To calculate the forecast for month 7, we have:

x
=

Standard Error of Estimate Example

S y,x

a y - b xy
n2

Example: Compute the Standard Error of the Estimate for our regression analysis example
S y,x

a y - b xy
n2

Correlation Coefficient Example

r

n x

n xy x y
2

x n y 2 y
2

Example: Compute the Correlation Coefficient of the data in our regression analysis example

10

Introduction to Linear Regression

Author(s)
David M. Lane
Prerequisites
Measures of Variability, Describing Bivariate Data
Learning Objectives
1. Define linear regression
2. Identify errors of prediction in a scatter plot with a regression line
In simple linear regression, we predict scores on one variable from the
scores on a second variable. The variable we are predicting is called
thecriterion variable and is referred to as Y. The variable we are basing
our predictions on is called the predictor variable and is referred to as X.
When there is only one predictor variable, the prediction method is
calledsimple regression. In simple linear regression, the topic of this
section, the predictions of Y when plotted as a function of X form a
straight line.
The example data in Table 1 are plotted in Figure 1. You can see that
there is a positive relationship between X and Y. If you were going to

11

predict Y from X, the higher the value of X, the higher your prediction of
Y.
Table 1. Example data.
X

1.00

1.00

2.00

2.00

3.00

1.30

4.00

3.75

5.00

2.25

Figure 1. A scatter plot of the example data.

Linear regression consists of finding the best-fitting straight line
through the points. The best-fitting line is called a regression line. The
black diagonal line in Figure 2 is the regression line and consists of the
predicted score on Y for each possible value of X. The vertical lines from
the points to the regression line represent the errors of prediction. As
you can see, the red point is very near the regression line; its error of

12

prediction is small. By contrast, the yellow point is much higher than the
regression line and therefore its error of prediction is large.

Figure 2. A scatter plot of the example data. The black line consists of the
predictions, the points are the actual data, and the vertical lines
between the points and the black line represent errors of prediction.
The error of prediction for a point is the value of the point minus the
predicted value (the value on the line). Table 2 shows the predicted
values (Y') and the errors of prediction (Y-Y'). For example, the first
point has a Y of 1.00 and a predicted Y (called Y') of 1.21. Therefore, its
error of prediction is -0.21.
Table 2. Example data.
X

Y'

Y-Y'

(Y-Y')2

1.00

1.00

1.210

0.210

0.044

2.00

2.00

1.635

0.365

0.133

3.00

1.30

2.060

0.760

0.578

4.00

3.75

2.485

1.265

1.600

5.00

2.25

2.910

0.660

0.436

13

You may have noticed that we did not specify what is meant by "bestfitting line." By far, the most commonly-used criterion for the best-fitting
line is the line that minimizes the sum of the squared errors of
prediction. That is the criterion that was used to find the line in Figure 2.
The last column in Table 2 shows the squared errors of prediction. The
sum of the squared errors of prediction shown in Table 2 is lower than it
would be for any other regression line.
The formula for a regression line is
Y' = bX + A
where Y' is the predicted score, b is the slope of the line, and A is the Y
intercept. The equation for the line in Figure 2 is
Y' = 0.425X + 0.785
For X = 1,
Y' = (0.425)(1) + 0.785 = 1.21.
For X = 2,
Y' = (0.425)(2) + 0.785 = 1.64.
COMPUTING THE REGRESSION LINE
In the age of computers, the regression line is typically computed with
statistical software. However, the calculations are relatively easy, and
are given here for anyone who is interested. The calculations are based
on the statistics shown in Table 3. MX is the mean of X, MY is the mean
of Y, sX is the standard deviation of X, sY is the standard deviation of Y,
and r is the correlation between X and Y.
Formula for standard deviation
Formula for correlation
Table 3. Statistics for computing the regression line.
MX

MY

sX

sY

2.06

1.581

1.072

0.627

The slope (b) can be calculated as follows:

14

b = r sY/sX
and the intercept (A) can be calculated as
A = MY - bMX.
For these data,
b = (0.627)(1.072)/1.581 = 0.425
A = 2.06 - (0.425)(3) = 0.785
Note that the calculations have all been shown in terms of sample
statistics rather than population parameters. The formulas are the
same; simply use the parameter values for means, standard deviations,
and the correlation.
STANDARDIZED VARIABLES
The regression equation is simpler if variables are standardized so that
their means are equal to 0 and standard deviations are equal to 1, for
then b = r and A = 0. This makes the regression line:
ZY' = (r)(ZX)
where ZY' is the predicted standard score for Y, r is the correlation, and
ZX is the standardized score for X. Note that the slope of the regression
equation for standardized variables is r.
A REAL EXAMPLE
The case study "SAT and College GPA" contains high school and
university grades for 105 computer science majors at a local state
school. We now consider how we could predict a student's university
GPA if we knew his or her high school GPA.
Figure 3 shows a scatter plot of University GPA as a function of High
School GPA. You can see from the figure that there is a strong positive
relationship. The correlation is 0.78. The regression equation is
University GPA' = (0.675)(High School GPA) + 1.097
Therefore, a student with a high school GPA of 3 would be predicted to
have a university GPA of

15

University GPA' = (0.675)(3) + 1.097 = 3.12.

Figure 3. University GPA as a function of High School GPA.

ASSUMPTIONS