jwb133 / smcfcs

R package implementing Substantive Model Compatible Fully Conditional Specification Multiple Imputation
11 stars 7 forks source link

Help wanted: Error in Surv(originaldata[, timeCol], originaldata[, dCol]) : Time variable is not numeric #13

Closed katriha closed 4 years ago

katriha commented 4 years ago

Hi

Sorry to bother you at this time and I really need your help now. I am trying to combing multiple imputations with cox proportional hazard model and luckily read your paper and find your smcfcs package in r. However, when I import my data use your example coding similar formula. Error appears and I dont know how to deal with.

The code I type in r was:

cox<- read_xlsx("Cox.xlsx") ##read the dataset names(cox)=c("t","d","a","b","c") ##name all the columns in the dataset cox<-cox %>% drop_na(t) ##drop all lines where t is NA cox<-cox%>%drop_na(d) ##drop all lines where d is NA

imps <- smcfcs(cox, smtype="coxph", smformula="Surv(t,d)~'a' + 'b'+'c'", method = c("","","norm", "norm","norm"),predictorMatrix = NULL, m = 10, numit = 100)

t is my censoring or event time and are all numeric values like this 3.52 1.60 2.30 1.63 12.53 9.92 4.87 and d is 0 or 1 which is whether the event happened

The error appeared was:

Error in Surv(originaldata[, timeCol], originaldata[, dCol]) : Time variable is not numeric

I don't know why this happening since all my time values are already numeric. Could you please give me some advice or suggestions that I can try to do something with it. I will so appreciated if you have any hints about this situation.

Best wishes, kaixin