Random Forest items to replace

Question

I am running the example of MRF in R developed here. The following blocs work perfect:

set.seed(0)
data=matrix(rnorm(15*200),200,15)
#DGP
library(pracma)
X=data[,1:3]
y=crossprod(t(X),rep(1,3))*(1-0.5*I(c(1:200)>75))+rnorm(200)/2
trend=1:200
data.in=cbind(y,data,trend)

Let’s say we want to predict the last 50 observations. We can set up our oos_pos as follows:

oos_position = nrow(data.in)-50: nrow(data.in)

However, I find the following error

Error in commitee[b, ] <- rt.output$pred : number of items to replace 
is not a multiple of replacement length  

when running the code below.

Is someone able to replicate the following code chunk?

mrf.output = MRF(data = data.in,
                 y.pos = 1,
                 x.pos = 2:4,
                 S.pos = 2:ncol(data.in),
                 oos.pos = oos_position,
                 mtry.frac = 0.25,
                 trend.push = 4,
                 quantile.rate = 0.3,
                 B = 100)

Any tip?

Answer 1

I don't think oos_position is specified incorrectly - it seems to be specifying the first 150 rows, rather than the last 50. Try this instead:

mrf.output = MRF(data = data.in,
                 y.pos = 1,
                 x.pos = 2:4,
                 S.pos = 2:ncol(data.in),
                 oos.pos = 151:200,
                 mtry.frac = 0.25,
                 trend.push = 4,
                 quantile.rate = 0.3,
                 B = 100)