similar to: Distance matrix by cosine?

Hello! Can I use DBM databases (as they are used in Perl) for the import of large datasets in R? Thanks in advance for your help, Petra Steiner - --------------------------------------------------- Petra Steiner Arbeitsbereich Linguistik Universitaet Muenster Huefferstrasse 27 48149 Muenster

I don't know the Russell-Rao coefficient but maybe this will help: You can compute the number of times y and x are both 1 (that's what your function f does) by counting 1s in y*x, i.e., sum(y*x), aka t(y)%*%x. Ordinary matrix multiplication does this row-by-column, so if M and N are two binary matrices, M%*%t(N) will have (i,j) entry equal to the count of 1s in common to row i of M and

Dear all, below you find enclosed my message from January 9th and my program (attention: beginner). Thanks for both answers! a. However, as far as I know the cosine is not the same as the Pearson correlation (only in special cases). b. Reid Huntsinger's hint was very useful, however I had to transpose the matrix first, for I want to calculate the distance of the _rows_. Regards, Petra

my major is bioinformatics, and i'm trying to cluster ( agglomerate the closest pari of observations ) in R. i have already got my own similarities metric, but do not know how to clust it based on similarities instead of dissimilarities. since the help document of hierclust mentions the parameter "sim", which seems good to me, but it doesn't appear in the code of hierclust()

my major is bioinformatics, and i'm trying to cluster ( agglomerate the closest pari of observations ) in R. i have already got my own similarities metric, but do not know how to clust it based on similarities instead of dissimilarities. since the help document of hierclust mentions the parameter "sim", which seems good to me, but it doesn't appear in the code of hierclust()

I use hierclust (hierarchical clustering) in multiv package. In the documentation it is said that plclust (plotting a dendrogram) is available in S-plus. Can I find it anywhere (I have searched through http://lib.stat.cmu.edu/S and found only quotations of plclust in multiv) or is it only part of the S package (which I don't have)? Thanks --------------- Charles RAUX, Laboratoire

Dear R-help list members i would like to use R to produce clustering or partitioning of a dataset. I am trying to use the functions: - hierclust() of the package multiv -pam(), agnes() and fanny() of the package cluster But I cannot get any result because of lack of memory. Would you know any clustering function not to greedy in memory? I have tried to expand my memory limit with memory.limit()

Given a matrix B, where B=A'A, how can I find A? In other words, if I have a matrix B which I know is another matrix A times its transpose, can I find matrix A? Thanks, Mike

Hallo Friedrich, I have read your announcement, downloaded and installed it in R-0.61.3. But now I have problems :) > hc <- hierclust(dist(t(pollen)), method=3) Error in pmatch(x, table) : argument is not of mode character > hc <- hclust(dist(t(pollen))) > members(hc) Error in .Fortran("assgn", n = as.integer(n), nplus1 = as.integer(nplus1), : C/Fortran function not in

Hello, a. when I use hclust with the methods media, centroid, and mcquitty, and plot the results, the dendrograms have lines that are crossing each other. Is this ok? b. My next question refers to plot.dendrogram: How can I use parameters as "hang" or "cex" here? E.g. for st <- as.dendrogram(subtreeshc[[x]]) I would like to have something like this, where cex and hang

Dear R helpers, is there a function or way within R to solve A'A=S for A, where all matrices have p x p order and S is a variance-covariance matrix? Thank you, Ralf Engelhorn

It is not clear to me that one can. If the singular value decomposition of A is the triple product P d Q', then the singular value decomposition of A'A=S is Q d^2 Q'. The information about the orthonormal matrix P is lost, is it not? ********************************************************** Cliff Lunneborg, Professor Emeritus, Statistics & Psychology, University of Washington,

Hello to all, I am using ArcView 3.2 under Wine. For now I have been able to do what I need, except for the projection utility... it seems that I am missing some dll, but I have no clue which ones could be... If somebody knows the solution, please let me know. I am new to Wine, and have no clue how to search for the right dll... Thanks a lot to all of you and to this great Wine-project!!! Petra

Hello, I have a short question about the prcomp function. First I cite the associated help page (help(prcomp)): "Value: ... SDEV the standard deviations of the principal components (i.e., the square roots of the eigenvalues of the covariance/correlation matrix, though the calculation is actually done with the singular values of the data matrix). ROTATION the matrix of variable loadings

Dear everyone, I have looked all over the internet but I cannot find a way to solve my problem. In my data I want to sum a couple of variables. Some of these variables have NA values, and when I add them together, the result is NA dat <- data.frame( id = gl(5,1), var1 = rnorm(5, 10), var2 = rnorm(5, 7), var3 = rnorm(5, 6), var4 = rnorm(5, 3), var5 = rnorm(5, 8) ) dat[3,3] <- NA dat[4,5]

Hello everybody, Is there in R a so called box m-test for testing the equality of the variance/cov. matrix for checking on homoscedasticity? I could not find it among the traditional packages for multivariate statistics... Petra -- Petra Wallem Centro de Estudios Avanzados en Ecolog伱伃a & Biodiversidad (CASEB) Departamento de Ecolog伱伃a Facultad de Ciencias Biol伱伋gicas Pontificia

Hello every body, I am trying to do a factorial anova analysis following this model: model<-anova(lm(responsevariable~factorA*factorB)) model<-anova(lm(luz$dosel~luz$estado*luz$Bosque)) Df Sum Sq Mean Sq F value Pr(>F) estado 1 6931.1 6931.1 41.6455 7.974e-06 *** Bosque 1 36.6 36.6 0.2197 0.6456 estado:Bosque 1 36.6 36.6 0.2197 0.6456 Residuals

Dear all, which font do I have to choose to use German umlauts at x/ylab in plot? Do I have to change Rdevga; and how? Or should I use these Hershey vectors? I looked quite a long time for a solution, which should be easy, so I thought I'd better ask here. Maybe some other people have already solved this problem. Regards, Petra - --------------------------------------------------- Petra

Okay, admittedly I'm running an old version 2.0.5, but, why do I have to set the eXecute permission in Linux before Samba will let a Windows client see any files? Why would execute have anything to do with read and/or write? I would think I could set -rw-rw---- or -rw-rw-rw- and be just fine, but oh no, you get an empty list and cannot create new files until I change it to -rwxrwx--- or

Hi, all. I'm trying to cluster 12,500 objects using hclust from package mva. The distance matrix takes up nearly 600 MB. The distance matrix also needs to be copied when being passed to the fortran routine that actually does the clustering (it's modified during the clustering), so that's 1200 MB. I'm actually on a machine with 2.5 GB of memory (and nothing else running), so I