|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: calculating a loglikelihood in regress
At 11:14 PM 5/2/2007, Dan MacNulty wrote:
Dear all,
Does anyone know if it's possible to calculate a loglikelihood value
from regress? I tried to download and implement the command
"reglike", but I was unsuccessful. Perhaps someone is aware of an
up-to-date alternative, or can provide further instruction on how to
use the reglike procedure. Any insights would be much appreciated.
e(ll) and e(ll_0) are stored by regress. Long and Freese's -fitstat-
command (part of their spost9.ado package) provides one of many ways
for displaying this and other useful info.
. webuse auto
(1978 Automobile Data)
. quietly reg price mpg rep78 foreign
. fitstat
Measures of Fit for regress of price
Log-Lik Intercept Only: -647.799 Log-Lik Full Model: -637.096
D(65): 1274.193 LR(3): 21.405
Prob >
LR: 0.000
R2: 0.267 Adjusted R2: 0.233
AIC: 18.583 AIC*n: 1282.193
BIC: 998.976 BIC': -8.702
BIC used by Stata: 1291.129 AIC used by Stata: 1282.193
Another way:
. nestreg, lr quietly: reg price ( mpg rep78 foreign)
Block 1: mpg rep78 foreign
+----------------------------------------------------------------+
| Block | LL LR df Pr > LR AIC BIC |
|-------+--------------------------------------------------------|
| 1 | -637.0963 21.40 3 0.0001 1282.193 1291.129 |
+----------------------------------------------------------------+
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/
