=+52. Assets and sales, part 2. The analyst in Exercise 28 realized the data were in need of transformation because of the nonlinearity between the variables. Economists commonly take the logarithm of these variables to make the relationship more nearly linear, and she did too. (These are base 10 logs.) The dependent variable is LogSales. The conditions for regression inference now appear to be satisfied.

Dependent variable is: LogSales R-squared = 33.9%

s = 0.4278 with 79 - 2 = 77 degrees of freedom Variable Coefficient SE(Coeff) t-Ratio P-Value Intercept 1.303 0.3211 4.06 0.0001 LogAssets 0.578 0.0919 6.28 60.0001