[R] Errors using nlme's gls with autocorrelation

[zerfetzen]

Hi,
I am trying out a generalized least squares method of forecasting that
corrects for autocorrelation.  I downloaded daily stock data from Yahoo
Finance, and am trying to predict Close (n=7903).  I have learned to use
date functions to extract indicator variables for Monday - Friday (and
Friday is missing in the model to prevent it from becoming full rank).  When
I run the following code...

> library(nlme)
> MyModel <- gls(Close ~ Monday + Tuesday + Wednesday + Thursday,
	correlation=corARMA(p=2), data=MyData, method="ML")

...I get the following error...

Error in corFactor.corARMA(object) : 
  Calloc could not allocate (62457409 of 8) memory

...Does anybody know what I'm doing wrong?  I appreciate any help.  Thanks.
-- 
View this message in context: http://www.nabble.com/Errors-using-nlme%27s-gls-with-autocorrelation-tp17026417p17026417.html
Sent from the R help mailing list archive at Nabble.com.