Dear Stata users,
I have unbalanced panel data on companies over years and am trying to run a
nonlinear regression in Stata which has lagged variables as regressors. I
keep coming up with the r(480) message which says "starting values invalid
or some RHS variables have missing values." The code I have written is
below:
program nlnissim
version 8.1
if "`1'" == "?" {
global S_1 "A G1 G2 G3 G4 G5 D1 D2 D3 D4 D5"
global A=1
global G1=1
global G2=1
global G3=1
global G4=1
global G5=1
global G6=1
global G7=1
global G8=1
global G9=1
global G10=1
global D1=1
global D2=1
global D3=1
global D4=1
global D5=1
global D6=1
global D7=1
global D8=1
global D9=1
global D10=1
exit
}
replace `1' = $A*L.OIa + ($G1*L.ireal46 + $G2*L2.ireal46 +
$G3*L3.ireal46 + $G4*L4.ireal46 + $G5*L5.ireal46 + $G6*L6.ireal46 +
$G7*L7.ireal46 + $G8*L8.ireal46 + $G9*L9.ireal46 + $G10*L10.ireal46) +
$A*($G1*L2.ireal46 + $G2*L3.ireal46 + $G3*L4.ireal46 + $G4*L5.ireal46 +
$G5*L6.ireal46 + $G6*L7.ireal46 + $G7*L8.ireal46 + $G8*L9.ireal46 +
$G9*L10.ireal46 + $G10*L11.ireal46) + ($D1*L.imktg + $D2*L2.imktg +
$D3*L3.imktg + $D4*L4.imktg + $D5*L5.imktg + $D6*L6.imktg + $D7*L7.imktg +
$D8*L8.imktg + $D9*L9.imktg + $D10*L10.imktg) + $A*($D1*L2.imktg +
$D2*L3.imktg + $D3*L4.imktg + $D4*L5.imktg + $D5*L6.imktg + $D6*L7.imktg +
$D7*L8.imktg + $D8*L9.imktg + $D9*L10.imktg + $D10*L11.imktg)
end
nl nissim OIa ireal46 imktg
It is an 11 lag model and my guess is that the error is arising since not
all companies have all 11 lags. Is there a way to tell Stata to ignore
those observations or any other way to get this to run?
Thanks much.
Terra
*
* 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/
