# st: AIC and BIC criteria for factor analysis

 From "Silia Vitoratou" <[email protected]> To <[email protected]> Subject st: AIC and BIC criteria for factor analysis Date Tue, 29 Jun 2004 10:36:04 +0300

```Dear statalisters

I am a beginner in factor analitic procedures and I am having problems
in choosing the correct number of factors to be extracted.
After using the ML method, my text books say that I should calculate AIC
and BIC criteria and choose the model with the lowest values.
But stata doesn't seem to calculate them after factor command and I cant
find the appropriate formulas that link the "factor" outcome with those
criteria. I found various formulas such as :
AIC(q) 	= (n/{n - 1 - (2p+4q+5)/6 }) &khgr;2(q)- 2df   (n=sample size, p =
variables, q = number of factors)

Is that correct?Should I use it?

Good day

Silia

*
*   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/
```