Closed Jingyan-uth closed 4 months ago
Do you have ties in your event times (i.e., event times that are exactly equal)?
From: yishigrace notifications@github.com<mailto:notifications@github.com> Date: Wednesday, 04 Sep 2019, 9:47 PM To: drizopoulos/JMbayes JMbayes@noreply.github.com<mailto:JMbayes@noreply.github.com> Cc: Subscribed subscribed@noreply.github.com<mailto:subscribed@noreply.github.com> Subject: [drizopoulos/JMbayes] rocJM issue (#48)
Hi there, I have an issue when running rocJM after I got results from mvJointModelBayes. When I fit rocJM, I kept having Error as: "Error in (function (x, vec, rightmost.closed = FALSE, all.inside = FALSE, : 'vec' must be sorted non-decreasingly and not contain NAs"
And the following is the traceback: 8.stop("'vec' must be sorted non-decreasingly and not contain NAs") 7.(function (x, vec, rightmost.closed = FALSE, all.inside = FALSE, left.open = FALSE) { if (!identical(FALSE, is.unsorted(vec))) ... 6.mapply(findInterval, Q_points, split(times, fids)) 5.right_rows(newdata, newdata[[timeVar]], newdata[[idVar]], GK_points_postRE) 4.survfitJM.mvJMbayes(object, newdata = newdata2, idVar = idVar, survTimes = Thoriz, M = M) 3.survfitJM(object, newdata = newdata2, idVar = idVar, survTimes = Thoriz, M = M) 2.rocJM.mvJMbayes(JMFit, newdata = test, Tstart = 2, Dt = 4) 1.rocJM(JMFit, newdata = test, Tstart = 2, Dt = 4)
Could please help give some advise?
Best, Grace
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdrizopoulos%2FJMbayes%2Fissues%2F48%3Femail_source%3Dnotifications%26email_token%3DADE7TT2WFKFVA3JBPJVLUMTQIAGDZA5CNFSM4ITVZ5G2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HJLBPGA&data=02%7C01%7Cd.rizopoulos%40erasmusmc.nl%7Cab4ed01e62b84959284408d73170ac94%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C637032232315650417&sdata=8FCLYgVvmRDQXEqRizBhS1L0%2Bbi6MDgi4fbuuqF1syU%3D&reserved=0, or mute the threadhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FADE7TT6BBLSVEWBAZ3VIR7TQIAGDZANCNFSM4ITVZ5GQ&data=02%7C01%7Cd.rizopoulos%40erasmusmc.nl%7Cab4ed01e62b84959284408d73170ac94%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C637032232315660414&sdata=ZgrPAemsF5lvmbq6e3%2Fdsw1jG3uEikE9iCe4jAgG5FQ%3D&reserved=0.
Yes, the event times have ties. I found that the newdata in rocJM() was not sorted by time. It works using sorted data. Thanks!
And Just curious, although mvglmer() allows missing observations in longitudinal biomarkers, does the function mvglmer() requires that multiple longitudinal biomarkers put in have same number of subject response?
Best, Grace
Hi there,
I had the same error message when fitting a JM model. I had a lot of ties in the event times. Does JM handle time events?
Thanks for reading! Best,
Lisa
Try JMbayes2.
Hi there, I have an issue when running rocJM after I got results from mvJointModelBayes. When I fit rocJM, I kept having Error as: "Error in (function (x, vec, rightmost.closed = FALSE, all.inside = FALSE, : 'vec' must be sorted non-decreasingly and not contain NAs"
And the following is the traceback: 8.stop("'vec' must be sorted non-decreasingly and not contain NAs") 7.(function (x, vec, rightmost.closed = FALSE, all.inside = FALSE, left.open = FALSE) { if (!identical(FALSE, is.unsorted(vec))) ... 6.mapply(findInterval, Q_points, split(times, fids)) 5.right_rows(newdata, newdata[[timeVar]], newdata[[idVar]], GK_points_postRE) 4.survfitJM.mvJMbayes(object, newdata = newdata2, idVar = idVar, survTimes = Thoriz, M = M) 3.survfitJM(object, newdata = newdata2, idVar = idVar, survTimes = Thoriz, M = M) 2.rocJM.mvJMbayes(JMFit, newdata = test, Tstart = 2, Dt = 4) 1.rocJM(JMFit, newdata = test, Tstart = 2, Dt = 4)
Could please help give some advise?
Best, Grace