Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

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

From |
"Martin Weiss" <martin.weiss1@gmx.de> |

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

Subject |
st: RE: post estimation tests with areg |

Date |
Tue, 21 Sep 2010 19:46:39 +0200 |

<> One obvious issue: The type of -predict-ion you want must be specified as an option. Otherwise Stata thinks you want a -varname- "xb" and another -varname-, which would be one too much... ************* sysuse auto, clear areg mpg weight gear_ratio, absorb(rep78) predict yhat, xb predict residual, res ************* HTH Martin -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Benhoen2 Sent: Dienstag, 21. September 2010 19:15 To: statalist@hsphsun2.harvard.edu Subject: st: post estimation tests with areg Hello Stata-listers, I am a novice Stata user (of 2 weeks) with more questions than answers. I am using the areg command to efficiently control for ~2000 fixed effects variables in a regression that has 3-7 independent variables for ~ 110,000 cases. Although the regression itself is very useful (and very fast), I have been unsuccessful at finding a way to do the following post-estimation activities: 1) save predicted values and residuals: I have tried using predict xp [varname] and predict r [varname], respectively. Both generate the following error, "too many variables specified" 2) save standardized residuals (Though if I had un-standardized residuals I could calculate myself) 3) test for heteroskedasticity 4) produce VIF statistics among IV, and 5) produce leverage statistics ...essentially many of the post-estimation options from regress. Are there any programs out there to produce these? (A SSC search was unsuccessful) Should I be using another regress command entirely? Is regress the only way to get there? If it turns out that regress is the only way to handle this, any advice on a) how to efficiently create the 2000 fixed effects variables, and b) encourage the most efficient use of regress with these variables. Thanks, in advance, for any and all. Ben Berkeley Lab * * 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/ * * 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/

**References**:**st: post estimation tests with areg***From:*"Benhoen2" <benhoen2@earthlink.net>

- Prev by Date:
**Re: st: 2-stages cluster analysis** - Next by Date:
**st: RE: post estimation tests with areg** - Previous by thread:
**st: post estimation tests with areg** - Next by thread:
**st: RE: post estimation tests with areg** - Index(es):