How about regressing each independent variable with its matrix of
constraints separately and then gathering all the constraints together
to estimate the full model.
sysuse auto,clear
const drop _all
gen time = _n
tsset time
pdlconstraints 12 4 A
pdlconstraints 6 2 B
qui cnsreg price l(0/12).mpg, con(A)
matrix dispCns, r
forv i = 1/8 {
constraint define `i' `=r(cns`i')'
}
qui cnsreg price l(0/6).weight, con(B)
matrix dispCns, r
forv i = 9/12 {
local j = `i' - 8
constraint define `i' `=r(cns`j')'
}
const dir
cnsreg price l(0/12).mpg l(0/6).weight, const(1-12)
Hope this helps,
Scott
----- Original Message -----
From: Alexander Severinsen <[email protected]>
Date: Wednesday, February 8, 2006 2:36 am
Subject: st: PDL and cnsreg
> Dear Statalisters,
>
> I am trying to estimate a polynomial ditributed lag model (PDL) as
> proposed by McDowell(2004) in The Stata Journal vol.4 nr.2 p.180-189.
> McDowell suggest using the constrained OLS instead of the Almon
> method,both producing the exact same estimates, the former
> requiring less
> effort. However I am having problems because I want to set
> restrictionson more than one independent variable.
<snip>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
