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

From |
John Antonakis <john.antonakis@unil.ch> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: Interpretation of Curvilinear Effects |

Date |
Tue, 09 Jun 2009 22:00:47 +0200 |

Hi Richard: By plotting I meant:

Best, J. ____________________________________________________ Prof. John Antonakis Associate Dean Faculty of Business and Economics University of Lausanne Internef #618 CH-1015 Lausanne-Dorigny Switzerland Tel ++41 (0)21 692-3438 Fax ++41 (0)21 692-3305 Faculty page: http://www.hec.unil.ch/people/jantonakis&cl=en Personal page: http://www.hec.unil.ch/jantonakis ____________________________________________________ On 09.06.2009 21:32, Richard Williams wrote:

At 02:42 PM 6/9/2009, John Antonakis wrote:Hi: First, don't use stepwise regression--it is the plague; no worse. Many journals simply won't even review manuscript with such data-driven methods (unless used for a particular goal--ridge, least-angular regression). For instance, see:It may depend on how you package it. On the one hand, stepwiseregression is the work of Satan; on the other hand diagnostic testsare good. So, one might just run the -estat ovtest- command and,based on it, decide that some sort of non-linearity should be allowedfor in the model.Also, I agree that plotting is a good idea, but really, how muchdifferent is that than stepwise regression? Either way, you arebasically looking at the data and deciding what to do with it. Botheyeballing the data and stepwise regression have the potential to makeyour significance tests deceptive, because you are using knowledgegained from the data itself and hence potentially capitalizing onchance in building your model. I guess the human judgment aspect ofplotting appeals to me over stepwise, but I also think it raises someof the same problems and concerns.Also, I don't know why you would need stepwise regression to justifythe possible inclusion of an x^2 term in the model. You can oftenthink of good substantive reasons for a curvilinear relationship. Ittherefore seems reasonable to me to test whether an x^2 term belongsin the model. I would certainly rather have some a priori theory inthere about an x^2 term rather than just seeing if it happens to makeit into a stepwise regression.My own view is that mindless stepwise is indeed the work of Satan.However, I think stepwise is potentially useful as a diagnosticdevice, e.g. is there reason to believe that my model may have omittedimportant variables? Even so, I think it is better if you have clearalternative or rival hypotheses in mind and explicitly test themrather than go on a fishing expedition with sw.------------------------------------------- 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/

* * 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: Interpretation of Curvilinear Effects***From:*Richard Williams <Richard.A.Williams.5@ND.edu>

**References**:**st: Interpretation of Curvilinear Effects***From:*Christian Weiss <christian.weiss@nightberry.de>

**Re: st: Interpretation of Curvilinear Effects***From:*John Antonakis <john.antonakis@unil.ch>

**Re: st: Interpretation of Curvilinear Effects***From:*Richard Williams <Richard.A.Williams.5@ND.edu>

- Prev by Date:
**Re: st: RE: Computing and allocating time intervals in a wide dataset** - Next by Date:
**st: Adjusted R square** - Previous by thread:
**Re: st: Interpretation of Curvilinear Effects** - Next by thread:
**Re: st: Interpretation of Curvilinear Effects** - Index(es):

© Copyright 1996–2015 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |