Hi,
I would like to use a weighted lm model to reduce heteroscendasticity. I am
wondering if the only way to generate the weights in R is through the
laborious process of trial and error by hand. Does anyone know if R has a
function that would automatically generate the weights need for lm?
On Tue, Apr 29, 2008 at 6:20 AM, tom soyer [EMAIL PROTECTED] wrote:
Hi,
I would like to use a weighted lm model to reduce heteroscendasticity. I am
wondering if the only way to generate the weights in R is through the
laborious process of trial and error by hand. Does anyone know if R has
Thanks Kingsford! I will cc r-help.
varPower() works for me. I want to use lm because predict.gls does not give
lwr and upr values, and also the bptest and ncv.test only work with lm
models...
On 4/29/08, Kingsford Jones [EMAIL PROTECTED] wrote:
hi Tom,
Basically, a simple way to model
3 matches
Mail list logo