I want to use a specialized function to compute knots on the fly to pass to splines::bs within a call to nlme::lme.
The specialized function will compute knots from the x-axis variable (the x argument in a call to splines::bs). The syntax works with lm. But when I try it with lme, the following error is returned: Error in model.frame.default(formula = ~why + eks + toydat + ID, data = list( : invalid type (list) for variable 'toydat' Below is a simplified example showing the "desired syntax," which returns the error, and a workaround. I am mystified what inherently is wrong with the "desired syntax", since it works fine with stats::lm. set.seed(5) library(splines) library(nlme) NID<-5 toydat<-data.frame(eks=0:0:((2*NID)-1)) toydat$ID<-factor(rep(LETTERS[1:NID], each=2)) toydat toydat$why<-with(toydat, as.integer(100*sin((eks / 7 * pi)^2) + rnorm(eks)/10)) customKnotsFn<-function(some)3.5 print(toydat) print(summary(toydat)) print(customKnotsFn) # lm has no trouble: mylm<-lm(why~bs(eks,knots=customKnotsFn(some=toydat$eks)), data=toydat) print(mylm$call) # The "desired syntax" returns an error: lme(fixed=why~bs(eks,knots=customKnotsFn(some=toydat$eks)),random=~1|ID, data=toydat) # Removing the argument from customKnotsFn eliminates the error, but then customKnotsFn is useless for practical purposes: lme(fixed=why~bs(eks,knots=customKnotsFn()),random=~1|ID, data=toydat) # Workaround returns no error: mylme<- with(toydat, lme(fixed=why~bs(eks,knots=customKnotsFn(some=eks)),random=~1|ID)) print(mylme$call) Of course, with the workaround the resulting mylme does not know where the data came from. Why should a workaround be necessary? Is there something inherently misguided about the "desired syntax" (above)? Thanks for any insight Jacob Wegelin
sessionInfo()
R version 3.1.3 (2015-03-09) Platform: x86_64-apple-darwin10.8.0 (64-bit) Running under: OS X 10.7.5 (Lion) locale: [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8 attached base packages: [1] splines stats graphics grDevices utils datasets methods base other attached packages: [1] nlme_3.1-120 loaded via a namespace (and not attached): [1] grid_3.1.3 lattice_0.20-30 ______________________________________________ R-help@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.
