The article on BMA (Bayesian model averaging) presents most valuable tools for model selection, but I find one detail confusing in Example 1. In page 4 of RNews 5/2, second paragraph says that the probability of Time variable not being in the model is 0.445. It seems to me that the figure should be 1 - 0.445 = 0.555, because p!=0.445 is the prob. of Time variable being in the model. The plot in Fig.2 is in line with this, since the height of scaled PDF seems to be 0.445 and the black spike points to 0.555. Have I understood this correctly? Regards, Antti Pirjetä, PhD student, Helsinki School of Economics. Mail to: pirjeta@hse.fi Muutoksia sähköpostiosoitteissamme: @hkkk.fi-päätteiset osoitteet ovat muuttuneet muotoon @hse.fi. Opiskelijoiden (nyt @ky.hkkk.fi) osoitteen loppuosaksi muuttuu tammikuussa 2006 @student.hse.fi. Vanhoihin osoitteisiin lähetetyt viestit ohjautuvat muutoksen jälkeenkin perille. Changes in our e-mail addresses: @hkkk.fi has been replaced by @hse.fi. Students' e-mail address (@ky.hkkk.fi) will change in January 2006 to @student.hse.fi. E-mails sent to the old address will be redirected to the new one. _____________________________________________________________ [[alternative HTML version deleted]]
The article on BMA (Bayesian model averaging) presents most valuable tools for model selection, but I find one detail confusing in Example 1. In page 4 of RNews 5/2, second paragraph says that the probability of Time variable not being in the model is 0.445. It seems to me that the figure should be 1 - 0.445 = 0.555, because p!=0.445 is the prob. of Time variable being in the model. The plot in Fig.2 is in line with this, since the height of scaled PDF seems to be 0.445 and the black spike points to 0.555. Have I understood this correctly? Regards, Antti Pirjet??, Helsinki School of Economics. Mail to: pirjeta at hse.fi Muutoksia s??hk??postiosoitteissamme: @hkkk.fi-p????tteiset osoitteet ovat muuttuneet muotoon @hse.fi. Opiskelijoiden (nyt @ky.hkkk.fi) osoitteen loppuosaksi muuttuu tammikuussa 2006 @student.hse.fi. Vanhoihin osoitteisiin l??hetetyt viestit ohjautuvat muutoksen j??lkeenkin perille. Changes in our e-mail addresses: @hkkk.fi has been replaced by @hse.fi. Students' e-mail address (@ky.hkkk.fi) will change in January 2006 to @student.hse.fi. E-mails sent to the old address will be redirected to the new one.
> Date: Fri, 18 Nov 2005 13:26:42 +0200 > From: "[iso-8859-1] Pirjet佷 Antti" <Antti.Pirjeta at hse.fi> > To: r-help at stat.math.ethz.ch > Subject: [R] R-News 5/2, Bayesian Model Averaging, a detail > > The article on BMA (Bayesian model averaging) presents most valuable tools for model selection, > but I find one detail confusing in Example 1. In page 4 of RNews 5/2, second paragraph says that > the probability of Time variable not being in the model is 0.445. It seems to me that the figure > should be 1 - 0.445 = 0.555, because p!=0.445 is the prob. of Time variable being in the model. > The plot in Fig.2 is in line with this, since the height of scaled PDF seems to be 0.445 and the > black spike points to 0.555. Have I understood this correctly?Yes, you are absolutely right. Thanks for pointing this out. Adrian Raftery ------------------------------------------------------------------- Adrian E. Raftery Professor of Statistics and Sociology Director, Center for Statistics and the Social Sciences University of Washington, Box 354320 Phone: (206) 543-4505 Seattle, WA 98195-4320. FAX: (206) 221-6873 Web: http://www.stat.washington.edu/raftery -------------------------------------------------------------------