problem with stepwiseglm selected features pvalues

1 次查看（过去 30 天）
Theo
Theo 2015-2-22
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?

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Generalized Linear Regression 的更多信息

标签

产品

Community Treasure Hunt

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

Start Hunting!

Translated by