a) Using the data in Table 1, specify a linear functional form for the demand for Combination 1 meals, and run a regression to estimate the demand for Combo 1 meals.
According to the passage, we know that the Quantity of meals sold by Combination (Q) is related to the average price charged (P) and the dollar amount spent on newspaper ads for each week in 1998(A). The price will influence the quantity of demand with inverse relation, and ads may lead to change of demand with positive relation.
Household income and population in the suburb did not change enough to warrant inclusion in the demand analysis, so we don’t think these variables should be involved in the function. What’s more, prices charged by other competing restaurants during 1998 could not be provided, so we have to regard it as parts of error item.
In summary, we specify a linear functional form for the demand as:
Qd=f(P,A)=α + βP + γA + ε
And we will run a regression to estimate the demand with: Qd =f(P,A)=^α + ^βP + ^γA.
b) Using statistical software, estimate the parameters of the empirical demand function specified in part a. Write your estimated demand equation for Combination 1 meals. Using Excel2010, we get the parameters of the empirical demand function specified in Part a. The Summary Output Table is attached as appendix. From that table, we get:
^α=100626.05, ^β=-16392.65, ^γ=1.58.
So we estimate the demand equation as blow:
Qd= 100626.05-16392.65P +1.58A
c) Evaluate your regression results by examining signs of parameters, p-values (or t-ratios), and the adjusted R2.
According the Summary Output Table, we get that: p-values are all much smaller than 5%, and t-ratios are all larger than 2. This means that there is less than 5 in 100 chance that the true coefficient of