[R] Error message using boot.ci function

[David Winsemius]

On 3/28/20 12:25 PM, varin sacha via R-help wrote:
> Dear R-experts,
>
> Here below my "toy" reproducible example showing many warnings and an error message. What I am interested in is the error message.
> Of course I can access and view the results doing : summary(results$t) to give me an idea of what is going on. But I don't know how to correct/solve my problem. Any help to solve my problem would be highly appreciated.


I do suggest you look at:


str(Dataset)

str(newdata)


The problem is actually in your construction of Dataset. Do not quote 
object names next time.

-- 

David

>
> # # # # # # # # # # # # # # # # # # # ## # # # # # # # # # # # # # #
> bmi=c(23,43,21,23,45,65,45,11,12,13,23,34,NA,NA)
> glucose=c(NA,12,23,11,12,21,23,21,23,43,23,12,NA,23)
> crp=c(123,212,154,342,123,111,121,765,453,123,213,211,NA,NA)
> age=c(67,45,34,56,87,NA,NA,23,18,65,45,87,65,33)
> sex=c(0,1,1,0,1,0,1,0,0,1,1,1,NA,NA)
>
> Dataset=data.frame("bmi","glucose","crp","age","sex")
> newdata=na.omit(Dataset)
>
> #install.packages( "robustbase",dependencies=TRUE )
> #install.packages( "boot",dependencies=TRUE )
> library(boot)
> library(robustbase)
>
>   # function to obtain MSE
>   MSE <- function(data, indices, formula,method) {
> fastMM <- lmrob( crp ~ bmi+glucose+age+sex, data=newdata)
>      d <- data[indices, ] # allows boot to select sample
>      fit <- lmrob(formula, data = d, method="MM")
>      ypred <- predict(fit)
>     mean((d[["fastMM"]]-ypred)^2)
>   }
>
>   # bootstrapping with 1000 replications
>   results <- boot(data = newdata, statistic = MSE,
>                    R = 1000, formula = crp ~ bmi+glucose+age+sex,method="MM")
>
> str(results)
> boot.ci(results, type="norm" )
>
> summary(results$t)
> # # # # # # # # # # # # # # # # # # # # # # # # #
>
> ______________________________________________
> 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.