EXAMPLE 11.9 Predicting Profits from Sales and Assets
bbcg30
CASE 11.1 Our examination of the logarithm-transformed explanatory and response variables separately and then in pairs did not reveal any severely skewed distributions with outliers or potential influential observations. Outputs for the multiple regression analysis from Excel, JMP, SAS, and Minitab are given in Figure 11.6. Notice that the number of digits provided varies with the software used. Rounding the results to four decimal places gives the least-squares equation
540