[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
Richard Williams <Richard.A.Williams.5@nd.edu> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: RE: Rituals [was: Terminology for supposedly all-purpose summaries] |

Date |
Mon, 11 Oct 2004 10:09:03 -0500 |

At 10:37 AM 10/11/2004 -0400, Stas Kolenikov wrote:

If by that you mean researchers are primarily interested in the R^2, I wouldn't agree (at least not always). If I have X1, X2, and X3 in the (constrained) model, and I then add X4, X5, and X6 to the (unconstrained) model, I'm likely interested in testing the hypothesisAlso, I've seen people deriving F-statistics for testing an increase in R2 as you add variables to the model. So it is the R2 that is the "first principle" for them. The distinction between "measures" and

H0: B4 = B5 = B6 = 0

HA: At least one of the above betas does not equal 0.

If the null is rejected, one consequence will be that R^2 will be significantly larger in the unconstrained model. I may use a formula that uses the R^2s in the calculations of the F statistic, but that does not mean I am focused on R^2 or have any interest in it other than as a piece of information needed to compute what I really want. I could use equivalent formulas using, say, the error sums of squares that give the same results. In other words, I would say that I am primarily interested in whether any of the added variables have nonzero effects, and the change in R^2 is simply a byproduct that I may or may not choose to focus on. I discuss this on pp. 6-7 of

http://www.nd.edu/~rwilliam/stats1/x93.pdf

What I'm describing is (what I call anyway) an incremental F test. In Stata, you could get an equivalent result by using the -test- command, which does a Wald test.

-------------------------------------------

Richard Williams, Associate Professor

OFFICE: (574)631-6668, (574)631-6463

FAX: (574)288-4373

HOME: (574)289-5227

EMAIL: Richard.A.Williams.5@ND.Edu

WWW (personal): http://www.nd.edu/~rwilliam

WWW (department): http://www.nd.edu/~soc

*

* 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/

**Follow-Ups**:**Re: st: RE: Rituals [was: Terminology for supposedly all-purpose summaries]***From:*Stas Kolenikov <skolenik@gmail.com>

**References**:**st: RE: Rituals [was: Terminology for supposedly all-purpose summaries]***From:*"Nick Cox" <n.j.cox@durham.ac.uk>

**Re: st: RE: Rituals [was: Terminology for supposedly all-purpose summaries]***From:*Stas Kolenikov <skolenik@gmail.com>

- Prev by Date:
**st: RE: Re: Data manipulation: how to keep only consecutive obs in an** - Next by Date:
**Re: st: RE: Re: Data manipulation: how to keep only consecutive obs in an** - Previous by thread:
**Re: st: RE: Rituals [was: Terminology for supposedly all-purpose summaries]** - Next by thread:
**Re: st: RE: Rituals [was: Terminology for supposedly all-purpose summaries]** - Index(es):

© Copyright 1996–2017 StataCorp LLC | Terms of use | Privacy | Contact us | What's new | Site index |