Timeline for How do you interpret coxph() output for the stratified Prentice, Williams, Peterson extension of the Cox PH model with gap-times?
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 12, 2017 at 2:59 | comment | added | N. Anderson |
I've located Frank Harrell Jr.'s book on regression modeling strategies. Hopefully that will give me a good grasp on the rms package. It's a little intimidating at nearly 600 pages. The R documentation, in comparison, is a svelte 246 pages. Thanks again for your help @EdM. I've had a heck of a time finding help at my own institution, and yours is greatly appreciated.
|
|
| Dec 12, 2017 at 2:47 | vote | accept | N. Anderson | ||
| Dec 12, 2017 at 1:43 | comment | added | EdM |
cph in rms pretty much works like coxph except that you use strat() instead of strata(); cluster() constructs are allowed in the formula. Worth the effort to learn rms if you will continue to do regression analysis of any type.
|
|
| Dec 11, 2017 at 21:45 | comment | added | N. Anderson |
Briefly, regarding ship, differences were expected and can be explained fairly satisfactorily. The insect is univoltine (1 generation per year) and individuals from eggs laid later in the year develop more quickly.
|
|
| Dec 11, 2017 at 21:44 | comment | added | N. Anderson |
Thanks for your response @EdM! I've come across rms before, but I was not sure how to implement it for a multi-event model. For coxph() I had come across a tutorial by Amarim and Cai 2015 (International Journal of Epidemiology). I'll have to look into rms more and see if I can figure it out. For some of my single-event models that do not meet the PH assumption, I have used the package MRH and its estimateMRH() function. That seemed to work well, but, again, I'm not sure how to expand it to multi-event models or if it is even possible.
|
|
| Dec 9, 2017 at 17:27 | history | answered | EdM | CC BY-SA 3.0 |