issue with survfitJM

[huyendn]

I got this error when trying to fit the dynamic prediction
Error in contrasts<-(*tmp*, value = contr.funs[1 + isOF[nn]]) :
contrasts can be applied only to factors with 2 or more levels
I know this error occurs when the categorical only has 1 level. However, if we do prediction for one patient, a categorical could have only 1 level for that patient. In your example code (https://erandrinopoulou.github.io/EducationalCorner_JMpred/) , the data for the patient for prediction has some categorical with one level. Could you please help look into this?

[drizopoulos]

Check JMbayes2 —— Professor of Biostatistics Erasmus Medical Center Rotterdam The Netherlands

…

________________________________ Από: huyendn ***@***.***> Στάλθηκε: Thursday, August 10, 2023 2:57:10 PM Προς: drizopoulos/JMbayes ***@***.***> Κοιν.: Subscribed ***@***.***> Θέμα: [drizopoulos/JMbayes] issue with survfitJM (Issue #104) Waarschuwing: Deze e-mail is afkomstig van buiten de organisatie. Klik niet op links en open geen bijlagen, tenzij u de afzender herkent en weet dat de inhoud veilig is. Caution: This email originated from outside of the organization. Do not click links or open attachments unless you recognize the sender and know the content is safe. I got this error when trying to fit the dynamic prediction Error in contrasts<-(*tmp*, value = contr.funs[1 + isOF[nn]]) : contrasts can be applied only to factors with 2 or more levels I know this error occurs when the categorical only has 1 level. However, if we do prediction for one patient, a categorical could have only 1 level for that patient. In your example code (https://erandrinopoulou.github.io/EducationalCorner_JMpred/) , the data for the patient for prediction has some categorical with one level. Could you please help look into this? — Reply to this email directly, view it on GitHub<#104>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADE7TTZPCIH7KSIKN6OYT4LXUUVINANCNFSM6AAAAAA3L5PS6Y>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[huyendn]

Thanks for the fast response. I just tried predict function in JMbayes2 with jm object and still got the same message error.

predSurv <- predict(jointm1.bayes, newdata = newdata_5, process = "event",

•                 return_newdata = TRUE)
  

Error in contrasts<-(*tmp*, value = contr.funs[1 + isOF[nn]]) :
contrasts can be applied only to factors with 2 or more levels

[huyendn]

Hello,

Do you know what might possibly cause this issue?

Thank you

[zoezhang106]

Hi,

I encountered the same problem recently and solved by converting categorical variables to factors. You may also want to check if each of the factors is with 2 or more levels.

Hope this helps.

[ellenerickson]

Make sure in "newdata_5" you are defining your factor variables to have two levels even if you're predicting with only one level. For example, if you have 2 time points and your categorical variable is gender, define it as:

newdata_5$gender <- factor(c('female','female'), levels = c('female','male'))