[R] Simulations study not working entirely...

[David Winsemius]

On 10/21/19 9:40 AM, varin sacha via R-help wrote:
> Dear R-Experts,
>
> Here below my reproducible example working but not entirely (working). What I understand is that there is a problem of libraries library(hbrfit) and ... ? How can I make it work entirely, many thanks for your precious help.
>
> ########SIMULATION STUDY 3 variables with 10% outliers n=2000
> install.packages( "robustbase" )
> install.packages( "MASS" )
> install.packages( "quantreg" )
> install.packages( "RobPer" )
> install.packages("devtools")  library("devtools") install_github("kloke/hbrfit")


When I attempted to replicate your code, I deciced to issue both of 
these commands in hte line above on separate lines. If you entered as 
above there shoiuld have been an error because there needs to be a 
semicolon to separate more than one  distinct command on a single line.

> install.packages('http://www.stat.wmich.edu/mckean/Stat666/Pkgs/npsmReg2_0.1.1.tar.gz')
> install.packages( "RobStatTM" )
>
>
> library(robustbase)
> library(MASS)
> library(quantreg)
> library(RobPer)
> library(hbrfit)
>
> library(RobStatTM)
>
> n<-2000
>
> x<-runif(n, 0, 5)
>
> z <- rnorm(n, 2, 3)
>
> a <- runif(n, 0, 5)
>
> y_model<- 0.1*x - 0.5 * z - a + 10
>
> y_obs <- y_model +c( rnorm(n*0.9, 0, 0.1), rnorm(n*0.1, 0, 0.5) )
>
>
> fastMM <- lmrob( y_obs ~ x+z+a)
>
> Huber <- rlm( y_obs ~ x+z+a)
>
> Tukey <- rlm( y_obs ~ x+z+a, psi = psi.bisquare )
>
> L1 <- rq( y_obs ~ x+z+a, tau = 0.5 )
>
> fastTau <- FastTau(model.matrix(~newdata$x+newdata$z+newdata$a),newdata$y_obs)


At this point the compiler should emit an error since newdata has not 
been created:


fastTau <- 
FastTau(model.matrix(~newdata$x+newdata$z+newdata$a),newdata$y_obs)
Error in eval(predvars, data, env) : object 'newdata' not found

  --

David.


>
> HBR<-hbrfit(y_obs ~ x+z+a)
>
> DCML <-lmrobdetDCML(y_obs ~ x+z+a)
>
>   
> MSE_fastMM<-mean((fastMM$fitted.values - y_model)^2)
>
> MSE_Huber<-mean((Huber$fitted.values - y_model)^2)
>
> MSE_Tukey<-mean((Tukey$fitted.values - y_model)^2)
>
> MSE_L1<-mean((L1$fitted.values - y_model)^2)
>
> MSE_fastTau<-mean((fastTau$fitted.values - y_model)^2)
>
> MSE_HBR<-mean((HBR$fitted.values - y_model)^2)
>
> MSE_DCML<-mean((DCML$fitted.values - y_model)^2)
>
>
> MSE_fastMM
>
> MSE_Huber
>
> MSE_Tukey
>
> MSE_L1
>
> MSE_fastTau
>
> MSE_HBR
>
> MSE_DCML
>
> ###############
>
> ______________________________________________
> 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.