To decide whether Y_i = β₀ + β₁X + u_i or ln(Y_i)= β₀ + β₁X + u_i fits the data better, you cannot consult the regression R² because

Question

Quizplus · Accepted Answer

The reason we cannot consult the regression R² to decide between the two models is that the Total Sum of Squares (TSS) is measured in different units between the models. The first model has TSS measured in the original scale, while the second model has TSS measured on the logarithmic scale. Therefore, we cannot compare the R-squared values between the two models to determine which model fits the data better.

To Decide Whether Yi = β0 + β1X + Ui

To Decide Whether Y_i = β₀ + β₁X + U_i