Dear Stata list,
I would very much appreciate your help with the following estimation
problem. I am trying to estimate the three parameters a, b, and c in
the regression:
y(t) = a + b* sum(c^s*x(t-s)), where s ranges from 0 to t.
That is, the three first terms in the regression are:
y(1) = a + b*x(1)
y(2) = a + b* [x(2)+c*x(1)]
y(3) = a + b* [x(3)+c*x(2)+c^2*x(1)]
Hence, there is an additional RHS term associated with each ascending
LHS variable, y(t).
I assume that the nl or ml Stata routines would be appropriate here
since this is a nonlinear problem, but I am not sure how to set up
the problem (so that the parameter b and the sum involving c could be
identified?).
Is it possible to estimate this jointly somehow, or do I have to run
the regression for various fixed c and then pick the one with the
best fit? If so, does anyone have a suggestion on how to do this in
an efficient way?
Thank you very much for your help
Bjorn
*
* 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/