Answer-first summary for fast verification
Answer: Exponentiate the RMSE, because the RMSE on the log scale corresponds to the geometric mean of the squared errors in the original scale.
When the label variable is log-transformed, the RMSE calculated on the log scale does not directly represent the error in the original scale. To convert the RMSE back to the original scale, you need to exponentiate the RMSE. This is because the RMSE on the log scale corresponds to the geometric mean of the squared errors in the original scale, and exponentiating the RMSE gives you the geometric mean of the errors, which is a more interpretable measure of error in the original scale.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
In a regression problem, you have calculated the RMSE of your model's predictions on the log scale of the label variable. You want to convert the RMSE back to the original scale of the label variable. Which of the following steps should you take, and why?
A
Exponentiate the RMSE, because the RMSE on the log scale corresponds to the geometric mean of the squared errors in the original scale.
B
Take the square root of the RMSE, because the RMSE on the log scale corresponds to the arithmetic mean of the squared errors in the original scale.
C
Multiply the RMSE by the mean of the original label variable, because the RMSE on the log scale is proportional to the original scale.
D
Do not convert the RMSE, because it is already in the original scale of the label variable.