Re: [R] Errors Calculating MVN Likelihood of Time Series with AR(1) Errors

Peter, Thanks very much for the advice. It looks like you're exactly right and the above-cited document may not be exactly right. Instead this ( http://pic.dhe.ibm.com/i

Re: [R] Errors Calculating MVN Likelihood of Time Series with AR(1) Errors

On 06 Mar 2014, at 19:16 , peter dalgaard wrote: > constant. Adding to that, and MA process with a neighbor correlation on the > order of .9 is not possible. To paraphrase, you're doing essentially this: Sorry, dyslectic fingers syndrome: [...] _an_ MA process [...] is not possible. -- Pe

Re: [R] Errors Calculating MVN Likelihood of Time Series with AR(1) Errors

On 06 Mar 2014, at 16:56 , Robert J. Kindman wrote: > Hi all, > > I'm having trouble calculating the likelihood of a time series with AR(1) > errors. I am generating my covariance matrix according to page 2 of ( > http://cran.r-project.org/doc/contrib/Fox-Companion/appendix-timeseries-regressi

[R] Errors Calculating MVN Likelihood of Time Series with AR(1) Errors

Hi all, I'm having trouble calculating the likelihood of a time series with AR(1) errors. I am generating my covariance matrix according to page 2 of ( http://cran.r-project.org/doc/contrib/Fox-Companion/appendix-timeseries-regression.pdf), using the library mvtnorm and the multivariate normal den