#2) Suppose you are convinced that the log of Intel's quarterly sales is not covariance stationary (it has a unit root) and decide to model it using an ARII) model. You use: In (5.1...) 'in (sacs.-.) = I + 61 [In (Salem) In (shawl + e. Table 3 shows the results of that regression. Table 3 Quarterly Observations, January l985Deocmbcr 1999 rStatistic 3.0875 2.4620 rSmusuc' ' 0. 1 088 0.6624 0.4 506 Regression Statistics Rsquared 0.0946 Standard error 0.0758 Observations 60 Durbin Watson 1 .9709 Coefcient Standard Error lump: 0.03 52 0.0 I 14 Lag 1 0.3064 0. 1244 Autumn-relations of the Rosidual Lag Aucocomcladon Standard Error 1 0-0 140 0. I 29 1 2 0-0355 0.1291 3 0.0532 0.1291 4 0.2 125 0. I 29 l A) Analyze the results of the regression? B) What conclusion can we make about the autocorrelations in the residuals? l .6463 C) What does this suggest about the model? Are the coefficients statistically significant? 0) How can we interpret the estimated coefficients in the model? E) Suppose we wanted to use this model at the end of the fourth quarter of 1999 to predict Intel's sales for the first quarter of 2000. Let us say that t is the fourth quarter of 1999, so t -1 is the third quarter of 1999 and t +1 is the first quarter of 2000. In the third quarter of 1999, Intel's sales were $7,328 million, in the fourth quarter of 1999, Intel's sales were $8,212 million. What would be the prediction for sales in the first quarter of 2000 be