problem with stepwiseglm selected features pvalues

3 ビュー (過去 30 日間)
Theo
Theo 2015 年 2 月 22 日
I'm using the following line for stepwiseglm
glm_model = stepwiseglm(chosen_features,idx,'constant','Distribution','binomial','upper','linear','PEnter',0.1, 'PRemove',0.15,'Criterion','aic');
I've set the Penter and Premove so that I can choose the pertinent features, however when the model is trained the feature pvalues are not within range, in fact I have PVALUES LIKE 0.3 in my model.
The only other thing I can think of is that there is 70% to 30% prevalence between my o and 1 classes.
Anyone has an idea?

回答 (0 件)

カテゴリ

Help Center および File ExchangeRegression についてさらに検索

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by