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

From |
lschoele@rumms.uni-mannheim.de |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: AW: st: re: beta coefficients for interaction terms |

Date |
Sun, 21 Jun 2009 18:04:20 +0200 |

I think Lisa refer the section on standardized regression cofficients of that book, particularly to the second item on pg. 201 (English edition). That item states that one should not use b*s_x/s_y to create standardized regression coefficients in the presence of interaction terms. Based on Aiken/West 1991 (28-48) it is recommended that one should standardize all variables that are part of the interaction in advance. Hence, instead of coding . sysuse auto, clear . gen ia = head*length . reg mpg head length ia, beta you should code . sysuse auto, clear . egen shead = std(headroom) . egen slength = std(length) . egen smpg = std(mpg) . gen ia2 = shead*slength . egen sia2 = std(ia2) . reg smpg shead slength sia2 So I guess the second option gives me the right beta coefficients.

Best, Lisa Zitat von Martin Weiss <martin.weiss1@gmx.de>:

<> Well, this thread has produced three options for you to choose from, and you have heard all the pros and cons. Just to summarize: ************* sysuse auto, clear /*first option */ qui gen ia2 = head*length reg mpg head length ia2, beta /*second option */ qui{ egen shead = std(headroom) egen slength = std(length) egen smpg = std(mpg) replace ia2 = shead*slength egen sia2 = std(ia2) } reg smpg shead slength sia2 /*third option */ reg mpg shead slength sia2 ************* HTH Martin -----Ursprüngliche Nachricht----- Von: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von lschoele@rumms.uni-mannheim.de Gesendet: Sonntag, 21. Juni 2009 17:33 An: statalist@hsphsun2.harvard.edu Betreff: Re: st: re: beta coefficients for interaction terms I am sorry, but I am getting kind of confused here. I am not a pro in running regressions with interaction terms, I am sorry that I have to ask so many questions. To get the beta coefficients I have to run the regression .reg smpg shead slength sia2 If I run the regression .reg mpg shead slength sia2 I get the coefficients, that are not standardized? Am I right? Best, Lisa Zitat von Kit Baum <baum@bc.edu>:<> Martin said " so it is not telling you anything for which you need to estimate the regression to find out." So the point estimate is equal to ouput of other commands, bout how about the CI? The unconditional CI is larger: . means mpg Variable | Type Obs Mean [95% Conf. Interval] -------------+---------------------------------------------------------- mpg | Arithmetic 74 21.2973 19.9569 22.63769 | Geometric 74 20.58444 19.38034 21.86335 | Harmonic 74 19.92318 18.81185 21.17405 ------------------------------------------------------------------------ because the standardized regressors that are added to the model are indeed correlated with mpg: _cons | 21.2973 .4121741 51.67 0.000 20.47524 22.11935 but as you can see that doesn't make much difference in testing any meaningful hypothesis about the mean of mpg. Kit Baum | Boston College Economics & DIW Berlin | http://ideas.repec.org/e/pba1.html An Introduction to Stata Programming | http://www.stata-press.com/books/isp.html An Introduction to Modern Econometrics Using Stata | http://www.stata-press.com/books/imeus.html * * 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/ * * 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: re: beta coefficients for interaction terms***From:*Kit Baum <baum@bc.edu>

**Re: st: re: beta coefficients for interaction terms***From:*lschoele@rumms.uni-mannheim.de

**AW: st: re: beta coefficients for interaction terms***From:*"Martin Weiss" <martin.weiss1@gmx.de>

- Prev by Date:
**st: AW: Listing data in a user-defined layout** - Next by Date:
**st: RE: AW: Rank transformation** - Previous by thread:
**AW: st: re: beta coefficients for interaction terms** - Next by thread:
**Re: st: re: beta coefficients for interaction terms** - Index(es):

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