Dear all,
I estimated a rational addiction model of this type:
y(t) = y(t-1) + y(t+1) + p(t),
where y is consumption and p are prices.
As my dataset consits of a panel of 18 cross and 36 monthly observations, i tried to estimate this model in first difference via ivreg2 (with gmm option), instrumenting lagged and future consumption with a set of past and future prices. The problem is that, while the test for overidentifying restrictions tells me that the full set of instruments is exogenous (and this is confirmed by checking the validity of each instrument with difference in Sargan test), abar test on residual displays covariance of first order in residuals.
Is this a big problem?
And why can I solve it?
Please replay as i've just tried (unsuccessfully) to include in the set of instruments also lagged and future consumption, as with monthly observations, i don't find a significant correlation with variables instrumented (confirmed by Sargan test that rejects the exogenous hypothesis of the full set of instruments)
Best regards
Andrea Sisto
____________________________________________________________
6X velocizzare la tua navigazione a 56k? 6X Web Accelerator di Libero!
Scaricalo su INTERNET GRATIS 6X http://www.libero.it
*
* 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/
