I am trying to estimate compensating variation using two-level NLOGIT
results. The difficulty is that I'm not sure if I have to rescale some
of the coefficients reported by Stata by multiplying by the applicable
inclusive value coefficient in order to use the formula for expected
maximum utility in McFadden or Kling and Thomson (1996 p.105).
Using the formulation in the stata manual
V= y*a+x*b where attributes y vary across first level alternatives and
x varies across the bottom level choices.
ivc=applicable inclusive value coefficient
Then the formula for
Maximum expected utility= Sum over 1st level alternatives[Sum over
bottom level alternatives(exp(V/ivc)]^(ivc)
Do stata folks know whether for the V term I can just sum the two linear
indexes given by stata, or do I have to rescale the coefficients (a) by
multiplying through by the iv term ?
Thanks in advance.
Bo Cutter
*
* 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/