Dear Statalisters,
I wrote the following simple ado file (as a pure exercise of
-ml-) to run an OLS model via -ml-:
/*----------olsviaml.ado----------*/
program olsviaml
version 8.0
args lnf theta1
quietly replace `lnf' = ln(normden($ML_y1-`theta1'))
end
/*----------olsviaml.ado----------*/
Then I ran the following codes,
. sysuse auto
. ml model lf olsviaml (foreign = mpg weight)
. ml maximize
and got the following results:
Number of obs = 74
Wald chi2(2) = 5.75
Log likelihood = -72.853871 Prob > chi2 = 0.0563
------------------------------------------------------------------------------
foreign | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mpg | -.0194295 .0342701 -0.57 0.571 -.0865977 .0477387
weight | -.0004678 .0002551 -1.83 0.067 -.0009678 .0000322
_cons | 2.123506 1.430791 1.48 0.138 -.6807933 4.927805
------------------------------------------------------------------------------
On the other hand, when running OLS via -regress- as
. reg foreign mpg weight
the results are as follows:
Source | SS df MS Number of obs = 74
-------------+------------------------------ F( 2, 71) = 21.05
Model | 5.75462023 2 2.87731012 Prob > F = 0.0000
Residual | 9.70483923 71 .136687876 R-squared = 0.3722
-------------+------------------------------ Adj R-squared = 0.3546
Total | 15.4594595 73 .211773417 Root MSE = .36971
------------------------------------------------------------------------------
foreign | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mpg | -.0194295 .0126701 -1.53 0.130 -.044693 .005834
weight | -.0004678 .0000943 -4.96 0.000 -.0006558 -.0002797
_cons | 2.123506 .5289824 4.01 0.000 1.068745 3.178267
------------------------------------------------------------------------------
The two results are virtually the same, but slightly different
as well; I'm afraid estimation results by -ml- are less
significant (ex.z-value vs.t-value, chi2 vs.F or R-squared),
though they may not be directly comparable.
So my question is, (suppose the above ado file is correct)
how one should interpret these two results (esp.the one by
-ml-).
Any suggestions welcome.
Thanks in advance.
K.I.
*
* 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/