Calculation Of Regression For Estimation Of Sales

Question

Data for corporate profits (P) and sales (S) for the manufacturing sector for the period

1970-2005 are stored in file sales.wf1.

a. Find least-squares estimates of β₁ and β₂ in the model

P = β₁ + β₂S + e

Report and comment on the estimated equation.

b. Draw a histogram of the residuals. Test for the normality of the error terms.

c. Use the Durbin-Watson test to test for positively-correlated first-order autoregressive errors. (Use a 5% significance level.)

d. Re-estimate the model assuming the existence of AR(1) errors. Report the results. Does ignoring the existence of AR(1) errors produce misleading inferences about β₁

and β₂?

e. Forecast profits for 2006 and 2007 assuming sales in these periods are 2400 and 2500, respectively. Find two sets of forecasts, one that assumes the errors do not follow an AR(1) process, and one that assumes that they do.

Summary

The question belongs to Statistics and it discusses about calculation of regression for estimation of sales.

Total Word Count 322