[R] Error in robdist.hbrfit(x) : x is probably collinear

Bert Gunter bgunter@4567 @end|ng |rom gm@||@com
Mon Mar 30 22:00:26 CEST 2020


I know nothing about the packages in question, but do you know what
"collinear" means and how/why it can mess up model fitting? If no,
that may be the problem. See also "overfitting."

Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )

On Mon, Mar 30, 2020 at 12:10 PM varin sacha via R-help
<r-help using r-project.org> wrote:
>
> Hi,
>
> A google search does not give me any hint :=(
> Maybe somebody can help me to fix the error message I get : Error in robdist.hbrfit(x) : x is probably collinear
>
>
> # # # # # # # # # # # # # # # # # # # ## # # # # # # # # # # # # # #
> bmi=c(23,43,21,23,45,65,45,11,12,13,23,34,NA,NA,34,35,45,65,43,23,12,11,15,43,23,88,78,79,89,89,99,43,21,34,32,45,65,76,56,45,34,23,12,32)
> glucose=c(NA,12,23,11,12,21,23,21,23,43,23,12,NA,23,11,12,32,12,14,12,11,10,9,8,9,8,7,90,76,32,12,11,12,23,11,123,32,12,14,34,54,65,76,87)
> crp=c(123,212,154,342,123,111,121,765,453,123,213,211,NA,NA,32,123,213,145,143,123,132,143,165,176,181,123,87,567,342,123,143,132,143,234,345,32,123,132,143,345,321,543,231,123)
> age=c(67,45,34,56,87,NA,NA,23,18,65,45,87,65,33,23,65,43,23,43,12,132,56,76,87,98,78,76,56,78,54,34,56,76,99,12,45,34,65,76,87,98,97,64,53)
> sex=c(0,1,1,0,1,0,1,0,0,1,1,1,NA,NA,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,1,1,1,0,0,0,1,1,0,0)
>
> Dataset=data.frame( bmi,glucose,crp,age,sex)
> newdata=na.omit(Dataset)
>
> install.packages( "boot",dependencies=TRUE )
> install.packages("remotes")
> remotes :: install_github("kloke/hbrfit",force=TRUE)
> install.packages( "quantreg",dependencies=TRUE  )
>
> library(boot)
> library(hbrfit)
> library(quantreg)
>
>   # function to obtain MSE
>  MSE <- function(data, indices, formula) {
>     d <- data[indices, ] # allows boot to select sample
>     fit <- hbrfit(formula, data = d)
>     ypred <- fit$fitted.values
>    mean((d[["crp"]]-ypred)^2)
>  }
>
> # bootstrapping with 100 replications
> results <- boot(data = newdata, statistic = MSE, R = 100, formula = crp ~ bmi+glucose+age+sex)
>
> str(results)
> boot.ci(results, type="norm" )
> # # # # # # # # # # # # # # # # # # # ## # # # # # # # # # # # # # #
>
> ______________________________________________
> R-help using 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.



More information about the R-help mailing list