crr.fit caused by extracting column wheninput data set is atibble (#1)QHScrnomo::prostate.dat) instead of using thedata() functionrms is now the only package thatQHScrnomo“Depends” on (i.e., it will be loaded when this package is loaded)cmprsk andHmisc have been moved to the“Imports” field and are accessed internally with thepkg::func format. Therefore, users shouldnotexpect these packages to load withQHScrnomo (though theyare required to be installed).sas.cmprsk to print the basefailure probability (1 - exp(-cumsum(f$bfitj))when thetime argument is used instead of thesurvival probability (exp(-cumsum(f$bfitj))pred.crr function from the package. Thisfunction was not previously exported or used, and was highly duplicativeofpredict.cmprsk. Nevertheless, it is no longer in thepackage.addOffset4ModelFrame,nomo2.crr,pred2.crr,pred3.crr,predictDesign)failcode inpred.ci tofailcode=1 to matchcmprsk::crr andQHScrnomo::crr.fittime intenf.crr andpredict.crr totime=NULL; also set default arguments forg,cuts,xlab,ylab arguments ingroupcitrace argument totenf.crr tooptionally trace the process in the console. It still does itby default, but previously it always traced it with no option to turn itoff.groupci, theci argument allows theuser to control whether we work on the scale ofsurvival orfailure (default) probabilities. When the argument was set toFALSE, this would apply a1-value operation toboth the within-group cumulative incidence estimate (to convertto survival)and the group-level mean (ofx).However,x is supposed to be an arbitrary continuousvariable that we can assess calibration for, not necessarily aprobability. Therefore, the change made in this version is toonly apply the1-value transformation to thewithin-group cumulative incidence andnot the group level meanforx.x, simply pass1-x into the functioncall.a andb arguments (for the intercept and slope, respectively), aswell asab,xlim, andylimarguments to adjust the graph accordingly.predict.cmprsk,sas.cmprsk,pred.ci,tenf.crr,groupci,nomogram.crrtestthatoldUnclass utility(internal) function. It’s definition wasfunction(x) unclass(x), so we just useunclass(x) instead.groupci, the user is now able to supplyxlab = "" as an argument. Also removed the usage ofsingle in the function definition in favor ofdouble as the former is only to be used in the context of.C/.Fortran (per documentation), which is not the casehere.nomogram.crr, changed some arguments:NULL default forfailtime tomake it more clear that this is a required argumentnomogram.crr andnomogram.mk6. Moved shortutility functions to be housed inR/nomogram.R directly(nomo2.crr,is.category,getOldDesign,Design.levels,value.chk,axisf)R/ forGetlim,predictDesign (containsaddOffset4ModelFrame),VarcovAdding historical change-tracking here that was included in theprevious version’s NEWS file (2.2.0) (in reverse order), but in aMarkdown-friendly format: