Dear all, Does anyone know if it is possible to estimate restricted arima models using the arima command in R? For instance lets say I want arima(6,0,0) but where the first three ar parameters are restricted to zero, ie. ar(1)=ar(2)=ar(3) = 0, and only the last three are actually estimated. Any help will be much appreciated. george -- View this message in context: http://r.789695.n4.nabble.com/restrected-arima-models-tp4671847.html Sent from the R help mailing list archive at Nabble.com.
Hello,
Take a look at parameter 'fixed' of function arima. From the help page
for arima():
"fixed
optional numeric vector of the same length as the total number of
parameters. If supplied, only NA entries in fixed will be varied."
Example use ('fixed' has length 7 because we must count the intercept
term):
x <- ts(cumsum(rnorm(100)))
fit <- arima(x, order = c(6,0,0), fixed = c(0, 0, 0, NA, NA, NA, NA),
transform.pars = FALSE)
fit
Hope this helps,
Rui Barradas
Em 18-07-2013 16:14, george escreveu:> Dear all,
> Does anyone know if it is possible to estimate restricted arima models
using
> the arima command in R? For instance lets say I want arima(6,0,0) but where
> the first three ar parameters are restricted to zero, ie. ar(1)=ar(2)=ar(3)
> = 0, and only the last three are actually estimated.
> Any help will be much appreciated.
> george
>
>
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/restrected-arima-models-tp4671847.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> 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.
>
On 18/07/2013 16:14, george wrote:> Dear all, > Does anyone know if it is possible to estimate restricted arima models using > the arima command in R? For instance lets say I want arima(6,0,0) but where > the first three ar parameters are restricted to zero, ie. ar(1)=ar(2)=ar(3) > = 0, and only the last three are actually estimated.Yes.> Any help will be much appreciated.Have you read the help page? Someone helped you by writing it (and indeed, by providing the facility, which few ARIMA-fitting programs have). fixed: optional numeric vector of the same length as the total number of parameters. If supplied, only ?NA? entries in ?fixed? will be varied. ?transform.pars = TRUE? will be overridden (with a warning) if any AR parameters are fixed. E.g. x <- rnorm(100) arima(x, order=c(6,0,0), fixed = c(0,0,0,NA,NA,NA,NA)) Coefficients: ar1 ar2 ar3 ar4 ar5 ar6 intercept 0 0 0 0.0423 0.0977 0.1299 -0.1462 s.e. 0 0 0 0.1065 0.1075 0.1059 0.1323> george-- Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595