I am working on the survey logistic regression. More than ten predictor variables are identified based on a P-value of < 0.25 on bivariate analysis. However, when I try multivariate analysis the model significance test fits only with a few variables (<5). If we add more variables, the model significance level deteriorates. So, shall I drop these variables (for the sake of model significance) even though they are important variables?
Asked
Active
Viewed 12 times
0
-
2There are many posts here already about feature selection (and stepwise model building, which you are doing "by hand"), and the general advice is "not do it". For more specific advice you need to give details. Maybe start with https://stats.stackexchange.com/questions/223808/a-more-definitive-discussion-of-variable-selection – kjetil b halvorsen Feb 16 '24 at 02:07
-
Use of Stata has no bearing on your question. – Nick Cox Feb 16 '24 at 09:57