Q. Demand can be predicted with experimental data, cross-section data or time-series data. In this case, cross-section data appear in the Excel file. Soft drink utilization in cans per capita per year is related to six-pack cost, income per capita, as well as mean temperature across the 48 contiguous states in the United States.

1. Given data, please assemble (a) a multiple linear regression equation as well as (b) a log-linear (exponential) regression equation for demand by MS Excel.

2. Given the MS Excel output in question 1, compare the 2 regression equations' coefficient of determination (R-square), F-test as well as t-test. Which equation is a good (better) fit? Which among the equation shows stronger overall significance to calculate the future demand? Which among the equation will you choose for a better demand estimation? Illustrate answer in the language of statistics.

3. By your choice of question 2, interpret each coefficient of independent variable in the soft drink demand estimated equation.

4. By your choice of equation in question 2, Clarify how many cans/capita/year on soft drink should be for a state in which 6-pack cost=$2.45, Income/Capita=$36,500, as well as Mean Temp= 68°F?

5. By your choice of equation in question 2 also the numbers in question 4, Calculate the cost elasticity of demand as well as income elasticity. Observe whether the demand is elastic or inelastic as well as whether soft drink is necessity, normal good or luxury good.

6. Now neglect cost as well as temperature from regression equation then runs the regression again. Illustrate the Excel output of only one independent variable, profits, should a marketing plan for soft drinks be designed that relocates most canned drink machines into low-income neighborhoods? Please clarify your answer in the language of economics.