Question: (5) It is known that if a correlation is not significant, the best predictor for the dependent variable is not \[{Y}'\] from the regression equation. Rather it is \[\bar{Y}\] the mean of the sample values of the dependent variable. What is the best predicted value for the following problem?

Suppose the regression line for 8 pairs of data linking cost of advertising and number of products sold is \[{Y}'\] = 2.8378 \[X\] + 54.8219, where \[X\] is in $1000 units. Assume the products are heavy duty tractors. If \[{{r}_{xy}}\] = 0.647 and \[\bar{Y}\] = 60 tractors sold, how many tractors would be sold if $5000 were spent on advertising? (Assume that $5000 is in the scope of the sampled data.) Show your method.