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 |
"Lachenbruch, Peter" <Peter.Lachenbruch@oregonstate.edu> |

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

Subject |
st: RE: RE: correcting skewness of an indep variable |

Date |
Mon, 15 Mar 2010 14:26:30 -0700 |

The typical assumption in regression / glm is that the predictor variables are measured without error, so skewness would not be an issue, although the variance might be. I'm not sure if taking logs would help, especially if the skewness was due to clumping at some value (such as 0). If you were using an errors-in-variables model it might be. Tony Peter A. Lachenbruch Department of Public Health Oregon State University Corvallis, OR 97330 Phone: 541-737-3832 FAX: 541-737-4001 -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox Sent: Monday, March 15, 2010 9:18 AM To: statalist@hsphsun2.harvard.edu Subject: st: RE: correcting skewness of an indep variable Skewness of an independent variable (I prefer an alternative term such as 'predictor') is sometimes an inconvenience. I am not aware that symmetry of an independent variable is ever an assumption in modelling. In the case of logit there are plenty of situations in which response is conveniently taken to be a function of log predictor. I don't think the more complicated treatment of -lnskew0- is ever obviously essential. In any case, the whole terminology of a 'correction' is at best unfortunate here. Can you reproduce the error message below with one of Stata's standard datasets? It looks a very odd error, particularly because I don't think -lnskew0- makes any use of Stata's matrix language. Perhaps your copy of Stata is corrupted. . set trace on to see where the error comes from. Nick n.j.cox@durham.ac.uk Fabio Zona When does one need to correct the skewness of an independent variable? I have a logit regression and my indep variable is strongly skewed; do I need to correct this (by using lnskew0 )?? I tried to correct my variable, but Stata gave this error message (see below); what does it mean? why it gave this message? lnskew0 lnindepvar = indepvar invalid matrix stripe; .z_ r(198); * * 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: correcting skewness of an indep variable***From:*Fabio Zona <fabio.zona@unibocconi.it>

**st: RE: correcting skewness of an indep variable***From:*"Nick Cox" <n.j.cox@durham.ac.uk>

- Prev by Date:
**st: xtabond and cluster** - Next by Date:
**RE: st: correcting skewness of an indep variables** - Previous by thread:
**st: RE: correcting skewness of an indep variable** - Next by thread:
**st: Panel probit/logit with instrumental variable** - Index(es):