Re: [R] SE for R-squared

2014-01-13 Thread peter dalgaard
On 12 Jan 2014, at 21:36 , John C Frain wrote: > Have a look at > > http://davegiles.blogspot.ie/2013/10/more-on-distribution-of-r-squared.html > Or head directly for the Wikipedia page for the noncentral beta distribution (and/or noncentral F). Giles doesn't really do the non-null distribut

Re: [R] SE for R-squared

2014-01-12 Thread John C Frain
Have a look at http://davegiles.blogspot.ie/2013/10/more-on-distribution-of-r-squared.html John On 10 January 2014 20:32, Troels Ring wrote: > In R package "psychometrics" an estimate of SE of R squared of /sersq <- > sqrt((4*rsq*(1-rsq)^2*(n-k-1)^2)/((n^2-1)*(n+3))) with n sample size, > and

Re: [R] SE for R-squared

2014-01-11 Thread David Winsemius
On Jan 10, 2014, at 12:32 PM, Troels Ring wrote: > In R package "psychometrics" an estimate of SE of R squared of /sersq <- > sqrt((4*rsq*(1-rsq)^2*(n-k-1)^2)/((n^2-1)*(n+3))) with n sample size, > and k number of parameters if sample size greater than 60 is found. > Does anyone have a formul

[R] SE for R-squared

2014-01-10 Thread Troels Ring
In R package "psychometrics" an estimate of SE of R squared of /sersq <- sqrt((4*rsq*(1-rsq)^2*(n-k-1)^2)/((n^2-1)*(n+3))) with n sample size, and k number of parameters if sample size greater than 60 is found. Does anyone have a formula for smaller sample size or an exact formula? I have been