You are on page 1of 10

CASE STUDY OF SOFT DRINK DEMAND ESTIMATION

Demand can be estimated with experimental data, time-series data, or cross-section data. Sara
Lee Corporation generates experimental data in test stores where the effect of an NFL-licensed
Carolina Panthers logo on Champion sweatshirt sales can be carefully examined. Demand
forecasts usually rely on time-series data. In contrast, cross-section data is appear in Table 1. Soft
drink consumption in cans per capita per year is related to six-pack price, income per capita, and
mean temperature across the 48 contiguous in the United States.
Table 1

Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montan
Nebraska
Nevada
New Hampshire
New Jersey

Cans/Capita/ 6-Pack $
Income
Yr
Price
$/Capita
200
2.19
150
1.99
237
1.93
135
2.59
121
2.29
118
2.49
217
1.99
242
2.29
295
1.89
85
2.39
114
2.35
184
2.19
104
2.21
143
2.17
230
2.05
269
1.97
111
2.19
217
2.11
114
2.29
108
2.25
108
2.31
248
1.98
203
1.94
77
2.31
97
2.28
166
2.19
177
2.27
143
2.31

Mean
Temp. F
13
17
11
25
19
27
28
18
14
16
24
20
16
17
13
15
16
21
22
21
18
10
19
19
16
24
18
24

66
62
63
56
52
50
52
72
64
46
52
52
50
56
56
69
41
54
47
47
41
65
57
44
49
48
35
54

New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rohde Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Total
Mean

157
111
330
63
165
184
68
121
138
237
95
236
222
100
64
270
77
144
97
102
7594

2.17
2.43
1.89
2.33
2.21
2.19
2.25
2.31
2.23
1.93
2.34
2.19
2.08
2.37
2.36
2.04
2.19
2.11
2.38
2.31
105.72

15
25
13
14
22
16
19
20
20
12
13
13
17
16
16
16
20
15
19
19
861

158.2083333

2.2025

17.9375

56
48
59
39
51
82
51
50
50
65
45
60
69
50
44
58
49
55
46
46
2573
53.6041666
7

QUESTION 1
Estimate the demand for soft drinks using a multiple regression program available on your
computer.
Estimated Demand for soft drink:
QD = 514.267 242.971 Price + 1.224 Income 2.931 Temp
(4.120)
r2 =0.698

(0.804)
SSE=38.261

Where the numbers in parentheses are t-scores.

(-5.582)

Multiple Regression :
Demand of soft drink : constant + 6 - pack price + income per capita + mean temp +error
: 514.27 242.97 6 pack price + 1.22 income per capita + 2.93 mean.Temp

Dependent Variable: CAN


Method: Least Squares
Sample: 1 48
Included observations: 48
Variable
C
PRICE
INCOME
TEMP
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic

Coefficient
514.2669
-242.9708
1.224164
2.931228
0.698024
0.677435
38.26108
64412.06
-240.9536
33.90231

Std. Error
113.3315
43.52628
1.522613
0.711458

t-Statistic
4.537722
-5.582162
0.803989
4.120027

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

QUESTION 2

Interpret the coefficients and calculate the price elasticity of soft drink demand.

Prob.
0.0000
0.0000
0.4257
0.0002
158.2083
67.36719
10.20640
10.36233
10.26533
1.980543

Both temperature and price are statistically significant with expected signs while income is
insignificant in its effect on soft drink demand.
for the log-linear model 3.12.

Mean P

=105.72 / 48
= 2.2025

Mean Q

= 7594 / 48
= 158.2083

Q/P

= -242.97

Price elasticity ED = (Q/P) (Mean P/Mean Q)


ED = (-242.97) / ( 2.2025 / 158.2083 )
ED = ( - 3.38 ) elastic

Interpretation on Price Elasticity: Based on the calculated price elasticity, the consumption on
soft drink is price elastic in nature. This means that for a 1% increase in price will result in more
than 1% decrease in quantity demanded for soft drinks.

This point elasticity at the mean price and quantity across the states is in the elastic range, as
expected. These are market-level price elasticities, so no firm behaviour is directly implied by
this estimate. An elastic demand at the market level does imply elastic firm-level demand at
comparable prices, comparable price sensitivity, and the smaller quantities facing each firm.

1)

The coefficient for demand for soft drink and price of soft drink is inverse relationship.

2) The quantity demand for soft drink per capita will change in opposite direction as the price of
soft drink change.

3) Demand for soft drink will reduce by 242.97 when price of soft drink change in the opposite
direction or inverse direction.
4)

The coefficient for demand for soft drink and income and demand for soft drink and

5)

mean temperature is positively relationship.


The quantity demand of soft drink will change in same direction as the income and mean
temperature change. So that, demand for soft drink will increase by 1.22 when income per
capita increase, and demand for soft drink also will increase by 2.93 when mean temperature
increase.

QUESTION 3

Omit price from the regression equation and observe the bias introduced into the parameter
estimate for income.

Income elasticity
Q = 514.89 - 242.88P + 1.22Y + 2.92T
Income elasticity, Ey = Q/Y x Y/Q
= 1.22*(17.89/160.76)
= 0.14
LogQ = 1.06 - 3.19LogP + 0.22LogY + 1.11LogT
Income elasticity, Ey = 0.22
Interpretation on Income Elasticity: Based on the calculated income elasticity, a positive
income elasticity indicates that soft drink is a normal goods.

log QD= 0.16 + 1.72 log TEMP 0.152 log INCOME


R2 = 0.49
(5.96)

( 0.73)

SSE = 0.137
When the independent variable of Price is removed from the equation, the R-Squared value
drops from 0.66 to 0.47. Thus the strength of correlation falls under moderate range (0.4 to 0.6).
The variables have a low association with the dependent variable as only 47% in quantity
demanded are explained by the independent variables.

QUESTION 4
Now omit both price and temperature from the regression equation. Should a marketing plan for
soft drinks be designed that relocates most canned drink machines into low income
neighborhoods? Why or why not?

Dependent Variable: CAN


Method: Least Squares
Sample: 1 48
Included observations: 48
Variable
C
INC

Coefficient
254.5629
-5.371683

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

0.111849
0.092542
64.17440
189444.3
-266.8446
5.793010
0.020162

Std. Error
41.09082
2.231815

t-Statistic
6.195129
-2.406867

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

Prob.
0.0000
0.0202

158.2083
67.36719
11.20186
11.27983
11.23132
2.313418

Omitting both price and temperature yields a linear model as follows:


QD = 254.563 5.372Y
QD = 254.6 5.37 INCOME R2 = 0.11
( 2.11) SSE = 64.2
For the log-linear model, one obtains
QD = 4.47 0.552 INCOME R2 = 0.09
( 2.13) SSE = 0.18
No, a marketing plan should not be designed specifically to introduce canned soft drink
machines into low-income neighborhoods. And students should not offer the negative and
significant income parameter estimate above as their reason. The above regression does NOT
call for relocating canned soft drink machines away from low-income neighborhoods. The
regression coefficient on income has been biased downward by the omission of price and
temperature enough to make an insignificant factor appear negative and significant in its effect
on demand. This illustrates the critical importance of using analytical reasoning and demand
theory to correctly specify a regression model.

INCOME
350
300
250
Q

200

Linear (Q)

f(x) = - 5.23x + 254.32


R = 0.11

150
100
50
0
5

10

15

20

25

30

The graph above shows the weak relationship between Income and Quantity Demanded. Thus,
the marketing plan should not be designed based on the income per capita factor as it does not
strongly correlated with the demand of soft drink cans.
Whether they market the product at low income groups or otherwise, it will not affect the
quantity demanded that much. We strongly believe that the company should not design their
marketing plan to relocate most canned drink machines into low-income neighbourhood.
In addition, as some variables i.e. price and temperature were removed from the equation, it is
unwise to rely solely on income factor to design on marketing plan as there exists a bias.
Instead of wasting resources in trying to influence a variable that is weakly related to the
dependant variable, the company should focus on other variables such as pricing as the critical
component of their marketing plan. Since price is strongly related to Quantity Demanded, the
company can stimulate the demand for their soft drink by giving discounts and "buy one, free
one" (BOGO) promotions.
The best demand specification

PRICE
350
300
250

f(x) = - 311.85x + 845.67


R = 0.57

200

Q
Linear (Q)

150
100
50
0
1.8

1.9

2.1

2.2

2.3

2.4

2.5

2.6

2.7

For Price, the R-squared is 0.5683 which is within the 0.4 to 0.6 range. Hence it has moderately
strong correlation.

INCOME
350
300
250
Q

200

Linear (Q)

f(x) = - 5.23x + 254.32


R = 0.11

150
100
50
0
5

10

15

20

25

30

For Income, the R-squared is 0.1094 which is within the range of 0 to 0.2. This indicates a very
weak correlation.

TEMPERATURE
350
300

f(x) = 4.91x - 104.03


R = 0.46

250

200

Linear (Q)

150
100
50
0
30

40

50

60

70

80

90

For Temperature, the R-squared is 0.4555 which is within the range of 0.4 to 0.6. Hence it has
moderate strong correlation.
Conclusion: The best demand specification is to remove income per capita from the regression
equation as the variable has a low correlation to the equation.