-------- Original Message -------- Subject: AW: Path Analysis Date: Mon, 15 Dec 2008 10:54:13 -0800 (PST) From: Fränzi Korner <[email protected]> To: <[email protected]> References: <[email protected]> Hi Thiago Macek I did once a path analysis in R "by hand", i.e. typing in the equations as presented in Sokal and Rohlf and using the command "solve()" for finding the coefficients. However, you have to be very carful not to make a misstake. If I had to do a path analysis now, I would use WinBugs, although it might not be necessary to use MCMC for the estimation of the coefficients of a path analysis. However, it is just very convenient to specify the equations of the path analysis in WinBugs - and the results are correct (at least I got the same results in R and in WinBugs for my path analysis, and it needed me about 6 hours to do it in R and only 15 min to do it in WinBugs). Good luck Fränzi -----Ursprüngliche Nachricht----- Von: morphmet [mailto:[email protected]] Gesendet: Montag, 15. Dezember 2008 16:22 An: morphmet Betreff: Path Analysis -------- Original Message -------- Subject: Path Analysis Date: Mon, 15 Dec 2008 05:46:09 -0800 (PST) From: thimacek <[email protected]> To: [email protected] References: <[email protected]> Dear friends at Morphmet, I'm planning to use Path Analysis to see the importance of several factors that I believe to be determinants of scapular shape, more or less like what's been done in Caumul & Polly 2005. I've been working on the Path Diagram to use and other things for some time, but I can't get around to a way to carry out the analysis. I thought it might be possible to do it in R, but I haven't found any guidelines in the internet about that. Could anyone enlighten me as to whether it's possible to do it in R, and other programs and possibilities to do it? Thanks in advance, Thiago Macek G. Zahn Museu de Zoologia da Universidade de São Paulo Seção de Mastozoologia Av. Nazaré, 481, Ipiranga (55) (11) 2065-8105 -- Replies will be sent to the list. For more information visit http://www.morphometrics.org -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
