Re: [R] Not missing at random

; > Xmiss <- X > > Xmiss[ids] <- NA > > Xmiss > > ____ > From: Joshua Wiley > To: Blaz Simcic > Cc: r-help@r-project.org > Sent: Mon, June 6, 2011 10:34:38 PM > Subject: Re: [R] Not missing at random > > Hi Blaz, > >

Re: [R] Not missing at random

Hi Blaz, See below. x <- matrix(c(1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,3,3,3,4), nrow = 7, ncol=7, byrow=TRUE) matrix pMiss <- 30 percent of missing values N <- dim(x)[1] number of cases candidate <- which(x[,1]<3 | x[,2]<

[R] Not missing at random

Hello!   I would like to sample 30 % of cases (with at least 1 value lower than 3 - in the row) and among them I want to set all values lower than 3 (within selected cases) as NA (NMAR- Not missing at random). I managed to sample cases, but I don’t know how to set values (lower than 3) as N

Re: [R] Not missing at random

Hi: Try this: x <- matrix(c(rep(1:5, 9), rep(3, 3), 4), ncol = 7, byrow = TRUE) ## indices of x for which the value is less than 3 candidates <- which(x < 3, arr.ind = TRUE)# generates 18 candidates grabs <- sample(nrow(candidates), 15) # 15/49 ~ 0.3 x2 <- x# copy x x2[candidates

[R] Not missing at random

Hello!   I would like to sample 30 % of cases (with at least 1 value lower than 3) and among them I want to set all values lower than 3 (within selected cases) as NA (NMAR- Not missing at random). I managed to sample cases, but I don’t know how to set values (lower than 3) as NA.   R code: