1. Chicken Palace periodically offers carryout five-piece chicken dinners at special prices.

Let Y be the
number of dinners sold and X be the price. Based on the historical observations and calculations in
the following table, determine the regression equation, correlation coefficient and coefficient of
determination. How many dinners can Chicken Palace expect to sell at $3.00 each?

Observation Price (X) Dinners sold (Y)

1 $2.70 760
2 $3.50 510
3 $2.00 980
4 $4.20 250
5 $3.10 320
6 $4.05 480

Solution:

Correlation
Coefficient of
Observatio Price Dinners of Intercept
Determination, Slope (b) Y=a + bX
n (X) sold (Y) Coefficient, (a)
r2
r
1 2.7 760 -0.843 0.711 1,454.60 -277.628
 
2 3.5 510        
 
3 2 980        
 
4 4.2 250        
 
5 3.1 320        
 
6 4.05 480        
 
  3          
621.721

Correlation of Coefficient, r = -0.843

Coefficient of Determination, r2 = 0.711
a = 1,454.60
b = -277.628
The Regression Equation is,
Y = a + bX
Y = 1454.60 -277.628 X
Y = 1454.60 -277.628 * (3)

= 621.721

= 622 (Appx.)

Answer: The Regression Equation, Y = 1454.60 + (-277.628) X

Correlation of Coefficient, r = -0.843; which means there is a strong negative correlation between
unit sales of dinner and sell prices.
Coefficient of Determination, r2 = 0.711; The unit sales information in the past had influenced by
sell price (independent variable) 71 times out of 100 times.
There could be approx. 622 dinners expected to be sold if sell at $3.00 each.
2. The monthly demand for units manufactured by the Acme Rocket Company has been as
follows:

Month Units Moth Units

May 100 September 105
June 80 October 110
July 110 November 125
August 115 December 120
a. Apply Naïve 1 method to forecast from June to January.
b. Use Naïve 2 method to forecast from June to January.
c. Use simple moving average to predict the demands for August through January.
Consider n to be 3.
d. Apply weighted moving average to predict the demands for September through
January. Consider n to be 4 and weights of 0.50, 0.30 and 0.20, with 0.50 applying to
the most recent demand.
e. Use the exponential smoothing method with α= 0.2 to forecast the number of units for
June to January. Choose initial forecast for May judiciously.
f. Calculate the MAD of forecasting technique exponential smoothing with α= 0.2
g. Considering the past data, which forecasting technique will you recommend for next
year between Simple moving average and exponential smoothing method with α=
0.2?
h. Calculate the tracking signal as of the end of December for exponential smoothing
method with α= 0.2. What can you say about the performance of your forecasting
method?
a. Apply Naïve 1 method to forecast from June to January.

Solution:

Naïve 1: Forecast of Future = Actual of Past

Ft+1 = At

Month Actual Data Forecasted Data

May 100  

June 80 100

July 110 80

August 115 110

September 105 115

October 110 105

November 125 110

December 120 125

January   120
b. Use Naïve 2 method to forecast from June to January.

Solution:

Naïve 2: Ft+1 = C

Usually C is an average over a number of past periods.

Forecasted
Month Actual Data Forecasted Data
Data
May 100    

June 80 100.00  

July 110 90.00 90

August 115 96.67 90

September 105 101.25 90

October 110 102.00 102

November 125 103.33 102

December 120 106.43 102

January   108.13 108.125

c. Use simple moving average to predict the demands for August through January. Consider n
to be 3.

Solution:

Simple Moving Average/ n-periods Moving Average

Here, n = 3 months Moving Average

F t+1 = (A t + A t-1 + A t-2) / 3

Month Actual Data Forecasted Data

May 100  

June 80  

July 110  

August 115 96.67

September 105 101.67

October 110 110.00

November 125 110.00

December 120 113.33

January   118.33
d. Apply weighted moving average to predict the demands for September through January.
Consider n to be 4 and weights of 0.50, 0.30 and 0.20, with 0.50 applying to the most recent
demand.

Solution:

Weighted Moving Average / n-Periods’ weighted Moving Average

Forecast for September,

W1 * AAugust + W2 * AJuly + W3 * AJune + + W4 * AMay

= ---------------------------------------------------------------------
W1 + W2 + W3 +W4

0.50 * 115 + 0.30 * 110 + 0.20 * 80 + 0.50 * 100

= ---------------------------------------------------------------------
0.50 + 0.30 + 0.20 + 0.50

57.5 + 33 + 16 + 50
= ------------------------------ = 104.333
1.5

Forecast for October,

W1 * A September + W2 * A August + W3 * A July + + W4 * A June

= -----------------------------------------------------------------------------
W1 + W2 + W3 +W4

0.50 * 105 + 0.30 * 115 + 0.20 * 110 + 0.50 * 80

= ---------------------------------------------------------------------
0.50 + 0.30 + 0.20 + 0.50

52.5 + 34.5 + 22 + 40
= ------------------------------ = 99.333
1.5

Forecast for November,

W1 * A October + W2 * A September + W3 * A August + + W4 * A July

= --------------------------------------------------------------------------------
W1 + W2 + W3 +W4
 0.50 * 110 + 0.30 * 105 + 0.20 * 115 + 0.50 * 110
= ------------------------------------------------------------------------
0.50 + 0.30 + 0.20 + 0.50

55 + 31.5 + 23 + 55
= ------------------------------ = 109.667
1.5

Forecast for December,

W1 * A November + W2 * A October + W3 * A September + + W4 * A August

= --------------------------------------------------------------------------------------
W1 + W2 + W3 +W4

0.50 * 125 + 0.30 * 110 + 0.20 * 105 + 0.50 * 115

= ------------------------------------------------------------------------
0.50 + 0.30 + 0.20 + 0.50

62.5 + 33 + 21 + 57.5
= ------------------------------ = 116
1.5

Forecast for January

W1 * A December + W2 * A November + W3 * A October + + W4 * A September

= -----------------------------------------------------------------------------------------
W1 + W2 + W3 +W4

0.50 * 120 + 0.30 * 125 + 0.20 * 110 + 0.50 * 105

= ------------------------------------------------------------------------
0.50 + 0.30 + 0.20 + 0.50

60 + 37.5 + 22 + 52.5
= ------------------------------ = 114.667
1.5
e. Use the exponential smoothing method with α= 0.2 to forecast the number of units for June to
January. Choose initial forecast for May judiciously.

Solution:
Simple Exponential Smoothing
Ft+1 = F t +  e t

  = 0.2 (Smoothing Parameter),

 Error (et) = (Actual demand for period t) – (Forecast for period t)

Month At Ft

May 100  

June 80 100.00

July 110 80 + 0.2 * (80-100) = 76

August 115 110 + 0.2 * (110-76) = 116.80

September 105  115 + 0.2 * (115-116.80) = 114.64

October 110 105 + 0.2 *(105-114.64) = 103.07

November 125 110 + 0.2 * (110-103.072) = 111.39 

December 120 125 + 0.2 * (125-111.386) = 127.72

January   120 + 0.2 * (120-127.723) = 118.46 

f. Calculate the MAD of forecasting technique exponential smoothing with α= 0.2

Solution:

Mean Absolute Deviation (MAD)

 et

l
MAD = -----------
n

Actual
|et|
Month
Data
Forecasted Data et |et| -----
n

May 100       13.386

June 80 100.00 -20.00 20

 

July 110 76.00 34.00 34

 

August 115 116.80 -1.80 1.8

 

September 105 114.64 -9.64 9.64

 

October 110 103.07 6.93 6.93

 

November 125 111.39 13.61 13.61

 

December 120 127.72 -7.72 7.72

 

January   118.46    
 
 g. Considering the past data, which forecasting technique will you recommend for next year
between Simple moving average and exponential smoothing method with α= 0.2?

Solution:

MAD and RMSE of Simple Moving Average

et2
| et |
MSE = RMSE =
Month At Ft et | et | et 2 -----
-------- MSE
n
n

May 100         8.667 123.333 11.106

June 80              

July 110              

August 115 96.67 18.33 18.33 336.11      

Septembe
105 101.67 3.33 3.33 11.11      
r

October 110 110.00 0.00 0.00 0.00      

November 125 110.00 15.00 15.00 225.00      

December 120 113.33 6.67 6.67 44.44      

January   118.33            

|et| et2=616.6
             
=43.33 7
 MAD and RMSE of Simple Exponential Smoothing with α= 0.2:

et2
| et |
MSE = RMSE =
Month At Ft et | et | et 2 -----
-------- MSE
n
n

May 100         13.386 277.861 16.669

June 80 100.00 -20.00 20 400.00    

 

July 110 76.00 34.00 34 1156.00    

 

August 115 116.80 -1.80 1.8 3.24    

 

September 105 114.64 -9.64 9.64 92.93    

 

October 110 103.07 6.93 6.93 48.02    

 

November 125 111.39 13.61 13.61 185.23    

 

December 120 127.72 -7.72 7.72 59.60    

 

January   118.46          
 
  |et|= 93.7 et2= 1945.03    
       

Answer: From both type of error calculation through MAD and RMSE, we have found that Simple
Moving Average has least error than Simple Exponential Smoothing technique. So, we’ll
recommend the forecasting technique Simple Moving Average for the next year forecasting.
h. Calculate the tracking signal as of the end of December for exponential smoothing method
with α= 0.2. What can you say about the performance of your forecasting method?

Tracking Signals (TS) = RSFE / MAD

RFSE= TS=
Month At Ft et MAD
et RSFE/MAD

May 100     15.380 13.386 1.149  1.2

June 80 100.00 -20.00    

 

July 110 76.00 34.00    

 

August 115 116.80 -1.80    

 
Septembe
105 114.64 -9.64    
r  

October 110 103.07 6.93    

 

November 125 111.39 13.61    

 

December 120 127.72 -7.72    

 

January   118.46      
 

      et = 15.38      

Answer: The Tracking Signal of Simple Exponential Smoothing technique has a value of 1.2 and
since an unbiased Forecasting technique has a range between -3 to +3, which clearly indicates
that it is unbiased till now and we can use it for next six months’ forecasting.