13.48 A research was conducted to determine the yearly death rate caused by road accidents in a country by two types of vehicles and the gross domestic product

($ million). The following regression model was fitted to 80 observations:

yn = 0.2831 + 0.7941x1t 10.13272

+ 0.4867x2t 10.15362

+ 1.8153x3t 10.49522 d = 1.66 where yn = death rate x1 = number of registered motorcycles per year x2 = number of registered cars per year x3 = gross domestic product The numbers below the coefficients are the coefficient standard errors.

a. Define the autocorrelated errors. What method can be used to detect the autocorrelated errors?

b. Construct an appropriate test to determine whether the above model’s error terms are positively autocorrelated at 5% significance level.

c. Explain how to improve the accuracy of the test suggested in part

b, and what assumption have be make.

d. Suppose the sample size, n, increases to 100 and d = 1.96. What can be concluded from the same test used in part b?