search

rpietro / airwayDehiscence

Airway dehiscence project using the UNOS database
0 stars 0 forks source link

cox-model doesn't work #13

Closed mworni closed 12 years ago

mworni commented 12 years ago

I tried to run the cox model - but it doesn't work.

survival1 <- coxph(Surv(AA_FU_TIME_IN_YEARS, AA_SURVIVAL_STATUS) ~ AA_Steroid_IND + AA_IL.2_IND, ties="efron") Error in coxph.control(...) : unused argument(s) (ties = "efron")

mworni commented 12 years ago

survival1 <- coxph(Surv(AA_FU_TIME_IN_YEARS, AA_SURVIVAL_STATUS) ~ AA_Steroid_IND + AA_IL.2_IND, data=airwayDehiscence)

made some changes - this seems to work - is it correct?

rpietro commented 12 years ago

strange, it works on mine:

survival1 <- coxph(Surv(AA_FU_TIME_IN_YEARS, AA_SURVIVAL_STATUS) ~ AA_Steroid_IND + AA_IL.2_IND, ties="efron")

below will test for proportional hazards assumption

prop.assump1 <- cox.zph(survival1) print(prop.assump1) #display results for assumption rho chisq p AA_Steroid_IND -0.00379 0.144 0.70423 AA_IL.2_IND 0.03056 9.425 0.00214 GLOBAL NA 9.446 0.00889

trying to think of a reason, but nothing other than a possible Schwiizertüütsch detection by some German hardware in your laptop. Tony, does it run for you?

On Thu, Jun 14, 2012 at 6:51 PM, mworni < reply@reply.github.com

wrote:

I tried to run the cox model - but it doesn't work.

survival1 <- coxph(Surv(AA_FU_TIME_IN_YEARS, AA_SURVIVAL_STATUS) ~ AA_Steroid_IND + AA_IL.2_IND, ties="efron") Error in coxph.control(...) : unused argument(s) (ties = "efron")

Reply to this email directly or view it on GitHub: https://github.com/rpietro/airwayDehiscence/issues/13