I'm using nl and would like to test equality between two coefficients
after estimation. According to the Stata manual test is available after
nl but I can't seem to get it to work. See my simple example below:
nl(ksumma04=exp({b0}+{b1}*boyta+{b2}*stand))
Source | SS df MS
-------------+------------------------------ Number of obs =
1738
Model | 6.3693e+09 3 2.1231e+09 R-squared =
0.9183
Residual | 566726732 1735 326643.65 Adj R-squared =
0.9182
-------------+------------------------------ Root MSE =
571.5275
Total | 6.9360e+09 1738 3990821.75 Res. dev. =
26995.96
------------------------------------------------------------------------
------
ksumma04 | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
/b0 | 6.715197 .0407067 164.97 0.000 6.635357
6.795036
/b1 | .001318 .0000904 14.58 0.000 .0011408
.0014953
/b2 | .0224096 .0012817 17.48 0.000 .0198958
.0249234
------------------------------------------------------------------------
------
* (SEs, P values, CIs, and correlations are asymptotic approximations)
test boyta=stand
regressor boyta not found
r(111);
Any other suggestions on how to specify the command? Found an earlier
question in the Statalist archive that gave the following suggestion but
it doesn't work either.
test _b[b1]=_b[b2]
[b1] not found
r(111);
I have also tried:
. test _coef[boyta]=0
. test _b[boyta]=_b[stand]
. test b1=b2
Nothing works. Any suggestions on the right syntax or is there some
other problem?
/Lina Jonsson
*
* 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/
