
Why perfect predictors result in a p-value of 0.93 ?


Why perfect predictors result in a p-value of 0.93 ?

How do hazards and maximum likelihood estimates predict event rankings?

When can large odds ratios and perfectly separated data bite you?

What do log-losses and perfectly separated data have to do with hockey sticks?

P-values below a certain threshold are often used as a method to select relevant features. Advice below suggests how to use them correctly.