[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Richard Williams <Richard.A.Williams.5@ND.edu> |

To |
"statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>, statalist <statalist@hsphsun2.harvard.edu> |

Subject |
Re: st: Model selection using AIC/BIC and other information criteria |

Date |
Tue, 23 Jun 2009 21:20:28 -0500 |

At 06:07 PM 6/23/2009, kokootchke wrote:

Dear all,I have a model that says that the return or yield spread of a bondissued by a country depends non-linearly on the country'sprobability of default. If I assume that this probability of defaultfollows a logistic form, I get that the log spread depends linearlyon "stuff" which I take to be macroeconomic variables. To choose thebest model, I use AIC/BIC.One interesting fact I observe is that in some cases, I see thatboth AIC and BIC select a model that contains some variable X evenwhen a lot of data points are missing for that particular variable,which means I actually lose a lot of observations when I includesuch variable X.More specifically, I have: MODEL 1 regress log_spread a b c X estat ic which gives AIC = 915 then, MODEL 2 regress log_spread a b c estat ic which gives AIC = 1500but the OLS in model 1 uses 1200 observations while the OLS in model2 uses 2800 observations (because 1600 observations are missing invariable X)!!You would think that this would be because X is very relevant toexplain the spread, but in fact I see some cases when this variableis statistically insignificant!!

I'm guessing a fairer comparison would be nestreg, lr: reg log_spread (a b c) X

------------------------------------------- Richard Williams, Notre Dame Dept of Sociology OFFICE: (574)631-6668, (574)631-6463 HOME: (574)289-5227 EMAIL: Richard.A.Williams.5@ND.Edu WWW: http://www.nd.edu/~rwilliam * * 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/

**Follow-Ups**:**RE: st: Model selection using AIC/BIC and other information criteria***From:*kokootchke <kokootchke@hotmail.com>

**References**:**st: Analyze a subpopulation of survey data in Stata 10.1***From:*"Karadogan, Figen" <fo145502@ohio.edu>

**st: Model selection using AIC/BIC and other information criteria***From:*kokootchke <kokootchke@hotmail.com>

- Prev by Date:
**st: Interpreting Poisson output** - Next by Date:
**RE: st: Model selection using AIC/BIC and other information criteria** - Previous by thread:
**st: Model selection using AIC/BIC and other information criteria** - Next by thread:
**RE: st: Model selection using AIC/BIC and other information criteria** - Index(es):

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