Suppose You Are Interested in Explaining Variation in Monthly Ice

Essay

Suppose you are interested in explaining variation in monthly Ice Cream consumption (thousands of gallons)and that you estimate the regression function (standard errors in parentheses) $$\begin{array} { llll } \text { Ice } \widehat { Crea}m _ { t } = &247.93 +& 12.22 \text { Income } _ { t } + &217.39 \text { Average Temperature } _ { t } \\& (98.62) &(4.35) & (81.63) \\n = 192 \\R ^ { 2 } = .9287 \\\end{array}$$
a)How many years' worth of data do you have? How can you tell? Explain.
b)Do you suspect that autocorrelation might be present in this model? If so,what type? Why? Explain.
c)How would you use the Durbin-Watson test to determine whether autocorrelation is present? Explain.What type of autoregressive process does the Durbin-Watson test work for?
d)Suppose you know that the autocorrelation follows an AR(1)process.How would you use the Cochrane-Orcutt method to correct for the autocorrelation? Explain.
e)When using the Cochrane-Orcutt method,how many observations will you have in your final analysis? Why? Explain.