Dear all: Subjects were measured two times (t1 and t2) on different variables (v1 ... vn). Between t1 and t2 there was an experimental manipulation. I computed two PCAs for time-points t1 and t2. Is it possible to combine both PCAs in order to get only one set of eigenvectors? Due to the experimental manipulation measurement values on time points t1 and t2 changed for each subject. Many thanks Stefan --
Hi Stefan,>> Is it possible to combine both PCAs in order to get only one set of >> eigenvectors?Yes there is: statis() in the ade4 package is probably what you want. In short, it does a k-table analysis that will give you a common ordination/position. It also shows how each time-set deviates from that common position. RV-coeff. showing how well the two tables match each other &c. Excellent graphical summaries... The method is quite complex, but if you read French then there is some very good documentation on their site: http://pbil.univ-lyon1.fr/ADE-4 Here you will also find a couple of papers (in English) on the method. There's also a list. Stephane (Dray) replies to many of the questions and writes and speaks English well. HTH, Mark. Stefan Koenig wrote:> > Dear all: > > Subjects were measured two times (t1 and t2) on different variables (v1 > ... vn). > Between t1 and t2 there was an experimental manipulation. I computed two > PCAs for time-points t1 and t2. > > Is it possible to combine both PCAs in order to get only one set of > eigenvectors? Due to the experimental manipulation measurement values on > time points t1 and t2 changed for each subject. > > Many thanks > Stefan > -- > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > >-- View this message in context: http://www.nabble.com/How-to-combine-to-PCAs-tp17665060p17667990.html Sent from the R help mailing list archive at Nabble.com.
Let V = (v11, v12, ..., v1n, v21, v22, ..., v2n) be the matrix of data with 2*n columns. Now, you simply do PCA on this data matrix. See the following paper, which has a related example: Gabriel, K. R. and Odoroff, C. L. (1990). Biplots in biomedical research. Statistics in Medicine, 9, 469-485. Ravi. -----Original Message----- From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of Stefan Koenig Sent: Thursday, June 05, 2008 4:47 AM To: r-help at r-project.org Subject: [R] How to combine to PCAs Dear all: Subjects were measured two times (t1 and t2) on different variables (v1 ... vn). Between t1 and t2 there was an experimental manipulation. I computed two PCAs for time-points t1 and t2. Is it possible to combine both PCAs in order to get only one set of eigenvectors? Due to the experimental manipulation measurement values on time points t1 and t2 changed for each subject. Many thanks Stefan -- ______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
hi, hi all, There are several methods to analyse two (e.g. coinertia, procuste) or K tables (e.g. statis, mfa, gpa, etc.) in the packages ade4, FactoMiner, etc. Otherwise analyses on instrumental variables can be an interesting way to explore your dataset (e.g. within/between analyses, etc.) hope this help pierre Selon Stefan Koenig <st.koenig at gmx.net>:> Dear all: > > Subjects were measured two times (t1 and t2) on different variables (v1 ... > vn). > Between t1 and t2 there was an experimental manipulation. I computed two PCAs > for time-points t1 and t2. > > Is it possible to combine both PCAs in order to get only one set of > eigenvectors? Due to the experimental manipulation measurement values on time > points t1 and t2 changed for each subject. > > Many thanks > Stefan > -- > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >