Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: st : Problem of no enough observations


From   orozco@toulouse.inra.fr
To   statalist@hsphsun2.harvard.edu
Subject   st: st : Problem of no enough observations
Date   Fri, 16 Nov 2007 16:23:05 +0100

Dear Statalist,

I wonder why Stata tells me that I haven't enough observations to estimate my
system of equations with "NLSUR". ("cannot have fewer observations than
parameters")

In fact, I have a system of equations to estimate. One equation per good `i' :
(the coefficients I have to estimate are between the braces {}.)

dw`i' = {b`i'} *ddeflexp + {g`i'vs1} * dlnp1+ {g`i'vs2} *  dlnp2+ {g`i'vs3} * 
dlnp3 + {g`i'vs4} * dlnp4 + {g`i'vs5} *  dlnp5 + {g`i'vs6} * dlnp6 + {g`i'vs7}
* dlnp7 + {g`i'vs8} * dlnp8 + {g`i'vs9} * dlnp9 + {g`i'vs10} * dlnp10 +
{g`i'vs11} * dlnp11 + {g`i'vs12} * dlnp12 )

As you can see, I have 13 coefficients to estimate for each good `i'
(i=1,...,12)
I have 78 observations. The problem just comes because I think Stata calculate
the total number of coefficients(13 * 12 = 156), but in fact I have only 13 per
good...

What's disturbing is that, I estimate a quite similar system with "reg3" (with
and without the SURE option) and have no problem...

Does someone have an idea???

Thank you very much.

Valérie


*
*   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/



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