# st: RE: Likelihood Function

 From "Julian Fennema" <[email protected]> To <[email protected]> Subject st: RE: Likelihood Function Date Thu, 24 Jul 2003 08:31:34 +0100

```Chuntao,

The function -phi- is, as you say, defined for the standard normal. The
relationship between this -phi- and the probability distribution
function for a generalised normal, i.e. where sigma=/=1, is
(1/sigma)phi((y-x*beta)/sigma).
Note that the (1/sigma) comes from the differentiation of the normalised
cumulative density function Phi((y-x*beta)/sigma) with respect to
x*beta.
Take logs, and you're there.

Julian

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Lee Chuntao
Sent: 24 July 2003 06:13
To: [email protected]
Subject: st: Likelihood Function

Dear Listers,

In page 29 of  Maximum Likelihood Estimation with Stata (Gould and
Sribney 1999), the likelihood function for the linear regression model
is
written as:

lnL=SUM (ln (phi((y-x*beta)/sigma)) - ln(sigma))
where phi() is the standard normal PDF.

My Question is: How the last term, ln(sigma), comes to the likelihood
function?

Can you knidly give me some ideas?

Chuntao

*
*   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/

*
*   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/
```