Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Estimating a nonlinear model


From   Steve Dolado <sdolado@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Estimating a nonlinear model
Date   Fri, 29 Jun 2012 12:17:28 -0400

Hi,
I have a model of the following form

M(y,X)= u  (u is the error term, y is the "dependent"variable and X is
a vector of regressors

I cannot isolate y here, so I guess nl will not work here.
More specifically,  the model is quadratic in the error term, so I can write

y + A(X1)*u  + B(X2)*u^2 = 0

Hence, I can write the solution in the form M(y,X)= u  by solving a
quadratic equation

u =( - A(X1) + (A(X1)^2 - 4*B(X2)*y) ^ 0.5)/(2*B(X2))

or

u =( - A(X1) - (A(X1)^2 - 4*B(X2)*y) ^ 0.5)/(2*B(X2))

where X1 and X2 are subsets of X with possibly overlapping components


Any idea how to estimate this model would be appreciated

Best
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index