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

From |
Kit Baum <baum@bc.edu> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
st: Re: formula for AIC in Stata |

Date |
Wed, 26 Oct 2005 07:50:51 -0400 |

Dirk said

try to open the -arimafit.ado- file in your STATA program folder and have a

look for the formula. Why are you interested in that?

arimafit is a user-written routine, available from SSC. You may

ssc type arimafit.ado

and you will find the line

local aic `ll2'+2*`np'

where ll2 is minus twice the likelihood value and np is the number of parameters. In 1998, official Stata did not calculate the AIC. Now, it does, with the estat command. As another poster wrote,

AIC = -2*ln(L)+2*p

where ln(L) is the log-likelihood of the model and p is the number of

parameters.

Kit Baum, Boston College Economics

http://ideas.repec.org/e/pba1.html

On Oct 26, 2005, at 2:33 AM, statalist-digest wrote:

try to open the -arimafit.ado- file in your STATA program folder and have a

look for the formula. Why are you interested in that?

* * 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/

- Prev by Date:
**RE: st: Numerous t-tests** - Next by Date:
**RE: st: Numerous t-tests** - Previous by thread:
**st: Numerous t-tests** - Next by thread:
**st: Sample size calculation** - Index(es):

© Copyright 1996–2016 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |