Timeline for Multivariable survival analysis: adding another variable lowers the p value?
Current License: CC BY-SA 3.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 21, 2016 at 8:50 | comment | added | Theodor | Identify in what way? If you are after model selection, then p-value based selection is arguably one of the worse ideas. Then it really depends what you are after: a model which explains the mechanism of the data or a prediction model. Anyway, there is a huge amount of literature, course materials and tutorials on the internet about variable selection. | |
| Apr 19, 2016 at 13:50 | comment | added | alize | Thank you. So it does not make sense at all to feed two dependent variables into a Cox model? For example, if I have to identify a predictor from a number of variables (they are all dependent), what can I do? | |
| Apr 14, 2016 at 15:02 | history | answered | Theodor | CC BY-SA 3.0 |