<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=iso-8859-1" http-equiv=Content-Type>
<META content="MSHTML 5.00.2919.6307" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2>Dear all,</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2> I'm currently using R v0.90.0 with packages
"nlme" on Windows 98 Second Edition. I have an enquiry on using the generalised
least squares function 'gnls'.</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2> My data set "lx.sea" contains some seasonal
GDP with some independent variables. I want to fit a nonlinear least
square using restricted maximum likelihood estimation. There exists a
AR(4) correlation structure (only the 4th AR) in the error term. So I enter
the following commands:</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>>
gnls(ldgdp~+pc2+pc3+pc4+pc5+pc7+t,data=lx.sea,correlation=corARMA(p=4,q=0))</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>It results not only the 4th AR but also the first 3
order AR in the equation. So how can I fit my model with only the 4th AR?? Are
there any other function in R allow me to do that??</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Thanks for the help and look forward to your
reply.</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Yours,</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Vincent</FONT></DIV></BODY></HTML>