Goodness on test or train set?

Question

I split my data set before on train (80%) and test (20%) splits. Trained logistic regression model on the train set. Now, want to check the goodness of fit using the Chi-square likelihood omnibus test, on what data set I should apply it to test or train?

I don't understand, what do you want to test exactly? Which statement are trying to prove or disprove? — Erwan, Jun 30 '22 at 08:15

score 0 · Answer 1 · edited Jul 15 '22 at 16:08

0

You should apply to the Test data as you already trained your model with training data.

edited Jul 15 '22 at 16:08

oW_

6,254
4
28
45

answered Jul 15 '22 at 08:29

Keshav Kumar

9
3

Goodness on test or train set?

1 Answers1