# construct the demand estimation for soft drink consumption

Demand can be estimated with experimental data, time-series data, or cross-section

data. In this case, cross-section data appear in the Excel file. 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 states in the United States.

**QUESTIONS**

- Given the data, please construct the demand estimation for soft drink consumption in

the United States by

(1) a multiple-linear regression equation (10%), and

(2) a log-linear (exponential) regression equation (10%).

- Given the MS Excel output in Question 1, please compare the two regression

equations’ coefficient of determination (R-square), F-test and t-test. Which equation

is a good (better) fit? Which equation shows the stronger overall significance to

predict the future demand? Which equation will you choose as a better estimation for

quantity demanded? Which equation will you choose as a better estimation for

elasticities? Explain your answer in the language of statistics. (20%)