[R] Maximum likelihood estimation

[Peter Dalgaard]

Jurica Brajković wrote:
> Hello, 
>
> I am struggling for some time now to estimate AR(1) process for commodity price time series. I did it in STATA but cannot get a result in R.
>
> The equation I want to estimate is: p(t)=a+b*p(t-1)+error 
> Using STATA I get 0.92 for a, and 0.73 for b. 
>
> Code that I use in R is: 
> p<-matrix(data$p) # price at time t
> lp<-cbind(1,data$lp) # price at time t-1
>
>                                              
> mle <- function(theta) {
>   sigma2<-theta[1]
>   b<- theta[-1]
>   n<-length(p)
>   e<-p-lp%*%b
>   logl<- -(n/2)*log(sigma2)-((t(e)%*%e)/(2*sigma2))
>   return(-logl)
>   }
>
>
> out <- optim(c(0,0,0),mle, method = "L-BFGS-B",
>     lower = c(0, -Inf, -Inf),
>     upper = c(Inf, Inf, Inf))
>
> The "result" I get is: " Error in optim(c(0, 0, 0), mle, method = "L-BFGS-B", lower = c(0, -Inf,:L-BFGS-B needs finite values of 'fn'"
>
> Can somebody spot the mistake?
>
> Many thanks, 
>
> Jurica Brajkovic 
>
>   
As far as I can see, the first element of the vector of starting values
(sigma2) is 0 and you are taking log of it....

> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>   


-- 
   O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark      Ph:  (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)              FAX: (+45) 35327907