[R] Survival model error

David Winsemius dwinsemius at comcast.net
Mon Nov 6 18:06:45 CET 2017


> On Nov 6, 2017, at 5:45 AM, Meghna Govil <megov31 at yahoo.com> wrote:
> 
> Thanks David. Could you show me how to do that in my example ? 

Possibly:

kmsurvival_np <- survfit(Surv( tte, censored) ~ . , data=train1_na)

I say "possibly" because I don't know whether all the columns of `train1_na` were included in the matrix you constructed. If they weren't then you would need to do something like:

kmsurvival_np <- survfit(Surv( tte, censored) ~ <lots of names separated by "+"'s> , data=train1_na)


The formalism of `reg_func( Y , Xm)` Where Xm is a matrix is typical for various machine-learning types of procedures, but breaks the "process-names-in-an-environment model of typical R regression functions. R regression functions often omit "step-down" options so loved by beginning SAS users.


> 
> Thanks,
> Meghna
> 
>> On Nov 6, 2017, at 12:58 AM, David Winsemius <dwinsemius at comcast.net> wrote:
>> 
>> You should stop trying to use matrices on the RHS and using separate vectors to Surv. Instead use a data argument and have the names in your formula refer to column names. 
>> 
>>>> David
>> 
>> Sent from my iPhone
>> 
>>> On Nov 5, 2017, at 7:21 PM, Meghna Govil via R-help <r-help at r-project.org> wrote:
>>> 
>>> 
>>> Hi - Below is my code and then the error when I run the last line. 
>>> 
>>> time_np <- train1_na$tte
>>> event_np <- train1_na$censored
>>> 
>>> 
>>> X_np <- cbind(
>>> train1_na$AMT, 
>>> train1_na$DISCOUNT_AMT,
>>> train1_na$high_price_pcnt,
>>> train1_na$EM_RECEIVED,
>>> train1_na$DM_RECEIVED,
>>> train1_na$TXN_WITH_RINGCODE,
>>> train1_na$WEB,
>>> train1_na$clearance_pcnt,
>>> train1_na$bts_pcnt,
>>> train1_na$sales_pcnt,
>>> train1_na$holiday_pcnt,
>>> train1_na$TXN,
>>> train1_na$REDEEMED_REWARDS
>>> 
>>> )
>>> 
>>> # Kaplan-Meier non-parametric analysis 
>>> 
>>> kmsurvival_np <- survfit(Surv(time_np,event_np) ~ X_np)
>>> 
>>> Error in `[.default`(y, who, 1) : (subscript) logical subscript too long
>>> 
>>> Any ideas?  
>>> I have tried several things and still get this error.  
>>> 
>>> 
>>> Thanks,
>>> Meghna
>>> 
>>> ______________________________________________
>>> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>> 
> 

David Winsemius
Alameda, CA, USA

'Any technology distinguishable from magic is insufficiently advanced.'   -Gehm's Corollary to Clarke's Third Law



More information about the R-help mailing list