Price of natural gas. Refer to the annual data on natural gas price 1Yt2, Exercise 14.42 (p. 14-38). You fit the simple linear regression model, E1Yt2 = b0 + b1t, to the data for the years 2000 to 2020 1t = 1, 2,

c, 212.

a. Find and plot the regression residuals against t. Does the plot suggest the presence of autocorrelation? Explain.

b. Conduct the Durbin-Watson test (at a = .05) to test formally for the presence of positively autocorrelated regression errors. [Hint: For this small sample, any DW d-value less than 1 indicates the presence of positive autocorrelation.]

c. Comment on the validity of the inference concerning model adequacy in light of the result of part b.