Re: [R] Getting bootstrap statistic to work

(data.cov) data.eigen$values[1]/sum(data.eigen$values) } Jean r-help-boun...@r-project.org wrote on 08/12/2011 09:10:52 AM: > [image removed] > > [R] Getting bootstrap statistic to work > > Alex Olssen > > to: > > r-help > > 08/12/2011 09:14 AM > >

Re: [R] Getting bootstrap statistic to work

Hi, It's kind of weird for me to not see a return() statement in the function. Maybe try rolling your own, nonparametric bootstraps aren't all that bad. i.e. reps=1000 stat.holder=rep(NA,reps) ###Rep of NA's so you can pick up errors easily for(count in 1:reps) { bootsample=sample(data, n.data

[R] Getting bootstrap statistic to work

Hi R-help, I am trying to implement a nonparametric bootstrap to find the standard errors of a simple statistics - the ratio of two scalars. I am having difficulty getting boot() to work correctly. I code a function to create the ratio of the relevant scalars. theta(data, i). When I call the fun