Hi all, This sounds like a potential interesting use for reduced major axis analysis, no?
Best, Ricardo Araújo www.paleomoz.org <https://streaklinks.com/A92nz3nu7mdYqR-GrQmTz0Ft/http%3A%2F%2Fwww.paleomoz.org%2F> ᐧ On Fri, Mar 11, 2022 at 4:24 PM Romain David <r.da...@nhm.ac.uk> wrote: > Dear Alejandro, > > Thank you for your reply. If I understand correctly, this issue arises > because I use PGLS regression, whereas I should use a correlation test. > Then I have a question. Is this correct to jointly estimate Pagel's lambda > on x and y, compute the PICS using the modified tree and calculate the > correlation between the PICS? > > As a note, for theoretical reasons I expect these two parameters to be > correlated, and therefore I was a bit puzzled by the two different results > I got. > > Best, > > Romain > ________________________________ > From: Alejandro Gonzalez Voyer <alejandro.gonza...@iecologia.unam.mx> > Sent: 10 March 2022 17:16 > To: Romain David <r.da...@nhm.ac.uk> > Cc: R-phylo Mailing-list <r-sig-phylo@r-project.org> > Subject: Re: [R-sig-phylo] Interpretation of Y~X vs. X~Y and estimation of > Pagel's Lambda using PGLSR > > Hello Romain, > > Your question has to do with the difference between a correlation, which > is symmetrical in the sense that the correlation between x and y is the > same as that between y and x, and a regression analysis which is not > symmetrical, as you show with the results which cause you some concern. > Although PGLS analyses are to be interpreted as associations (or > correlations) between traits (given there is no way to “know” whether there > is indeed a causal relationship, see e.g. Martins 2000 among others), > mathematically they are fit as a linear model (a regression analysis), > where the difference with traditional OLS is that a degree of covariance in > the residuals is estimated by a given evolutionary model. Thus, the > analysis is asymmetrical and a decision, a priori, must be taken by the > researcher as to which variable is deemed to be the response and which is > the predictor, even though associations cannot be interpreted as causal. > > Cheers > > Alejandro > _______________________________________________ > Dr Alejandro Gonzalez Voyer > Investigador Titular B > > Coordinación de Docencia y Formación de Recursos Humanos > Instituto de Ecología > > Laboratorio de Conducta Animal > Instituto de Ecología > Circuito Exterior S/N > Ciudad Universitaria > Universidad Nacional Autónoma de México > México, D.F. > 04510 > México > > Tel: +52 55 5622 9044 > E-mail: alejandro.gonza...@iecologia.unam.mx<mailto: > alejandro.gonza...@iecologia.unam.mx> > > Web Site: > https://uk01.l.antigena.com/l/TTMJvOPtnUfbtSnveKCJTcfVemCxSYgvypa2pfttyL8g9xLbeB2b9iAUjxU8jWnbP02yy68t7iTMUxUAJe3hAHPddWk87Web2EkCjbkaMKk7Ik82Wn82bA80IUbjWmdddd > > > On 10 Mar 2022, at 4:32, Romain David <r.da...@nhm.ac.uk<mailto: > r.da...@nhm.ac.uk>> wrote: > > Dear all, > > I am running into an issue with the interpretation of two regressions > using PGLSR. Using the same sample, when I regress Y on X, I get a p-value > of 0.10 and a lambda of 0.9. However, when I regress X on Y, I get a > p-value << 0.001 and a lambda of 0.2. See the attached graphs. It seems > that it has to do with the estimation of lambda and the peculiar > phylogenetic distribution and spread of the residuals (e.g. grey dots vs. > yellow dots). Would you interpret that these two variables are related? Is > there a way to estimate a "joint" lambda so the two regressions are > reciprocal? > > Thank you very much. > > Romain > <RsigPhylo_Example.jpg>_______________________________________________ > R-sig-phylo mailing list - R-sig-phylo@r-project.org<mailto: > R-sig-phylo@r-project.org> > https://stat.ethz.ch/mailman/listinfo/r-sig-phylo > Searchable archive at > http://www.mail-archive.com/r-sig-phylo@r-project.org/ > > > [[alternative HTML version deleted]] > > _______________________________________________ > R-sig-phylo mailing list - R-sig-phylo@r-project.org > https://stat.ethz.ch/mailman/listinfo/r-sig-phylo > Searchable archive at > http://www.mail-archive.com/r-sig-phylo@r-project.org/ > [[alternative HTML version deleted]] _______________________________________________ R-sig-phylo mailing list - R-sig-phylo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-phylo Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/
