similar to: Loadings and scores from fastICA?

Hi all, I'm new to PCA in R, so this might be a basical thing, but I cannot find anything on the net about it. I need to make a PCA plot with two response variables (df$resp1 and df$resp2) against eight metabolites (df$met1, df$met2, ...) and I don't have a clue how to do... and I've only used the simplest PCAs before, like this: pcaObj=prcomp(t(df[idx, c(40:47)]))

Hi everyone It looks like repeated runs of fastICA produce quite significantly different mixing matrices (not only in terms of sign and row order). I'm not a specialist, so would appreciate any advice on whether this should really be the case: > res3 = > fastICA(af[,2:20],4,alg.typ="parallel",fun="logcosh",alpha=1,method="C",row.norm=TRUE) colstandard >

Hi all, I would like to remove rows from a matrix, based on the frequency of missing values. If there are more than 10 % missing values, the row should be deleted. I use the following to calculate the frequencies, thereby getting a new matrix with the frequencies: freqNA=rowMeans(is.na(exprdata)) But is there a shorter way to remove the rows based on "(1-freqNA)>0.1"

Hi all, I need to replace missing values in a matrix by 10 % of the lowest available value in the matrix. I've got a function I've used earlier to replace negative values by the lowest value, in a data frame, but I'm not sure how to modify it... nonNeg = as.data.frame(apply(orig.df, 2, function(col) # Change negative values to a small value, close to zero { min.val =

Hello, I think to use FastICA package for microarray data clusterization, but one question stops me: can I know how much variance explain each component (or all components together) ? I will be very thankful for the help. Thanks, Pavel [[alternative HTML version deleted]]

A colleague suggested that some R software was available for estimating "Independent Components Analysis" (ICA) (... signal separation). If so, I''d appreciate any pointers . Chris Barker Director of Statistical Research MEDTAP International, Inc. Redwood City, Ca www.medtap.com <http://www.medtap.com/> 650 632 4218

Hi, I'm trying to use fastICA package but I only get an error message like following. > library(fastICA) Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/usr/lib/R/library/fastICA/libs/fastICA.so': /usr/lib/R/library/fastICA/libs/fastICA.so: undefined symbol: sgesdd_ Error: package/namespace load failed for 'fastICA' Here's the

I have been playing around with the FastICA algorithm in R, and I have come across the following issue. I'm new to this topic, so I'm hoping someone will be able to shed some insight. We know that the variance of the Independent Components is constrained to be 1 and they are uncorrelated, so the sum of squares of the coefficients of the mixing matrix (a$A) will give the variance of the

Hi all, I'm doing Multiple linear regression for a data set. However, it takes a lot of time, as I would like to check every possible combination of factors, evalute the results based for instance on their p values, and then choose the best regression model. So, I wonder if anyone might have a script for that? Or if not, do you have some suggestions how to create such a script? I've

Hi all, I'm have a matrix (X) with observations as rows and parameters as columns. I'm trying to exchange all missing values in a column by the column mean using the code below, but so far, nothing happens with the NAs... Can anyone see where the problem is? N<-nrow(X) # Calculate number of rows = 108 p<-ncol(X) # Calculate number of columns = 88 # Replace by columnwise

Hi all, I'm making a PCA plot with eight variables (columns taken from a larger data frame "fieldTrial0809[idx,c(39:46)]"). I then want the symbols in the plot to be colored as a gradient from red to blue, depending on the value of another column in "fieldTrial0809[idx, c(48)]" containg temperatures from -12.1 to -5.4. I don't want to use the heat.colors(n)

Hi all, I guess this might be an easy question, but I've searched multiple help pages without finding any answear... so now I put my trust in you! I have a data frame (36 variables and 556 observations). One column contains three factors, and I would like to divide the data frame into three new ones, based on the value of the factors, thereby having only one value for all elements of the

Hi ALL, Recently I confronted with a error in using package ProDenICA, when let k<p (exactly the example code with k=1, http://finzi.psych.upenn.edu/R/library/ProDenICA/html/ProDenICA.html), Error in solve.default(V, W) : 'b' must be compatible with 'a' I have examined the code of function ProDenICA(), but could not find the reason. And how large is the sample size N

Hi, Looking at the fastICA library. I want to test separating out different sounds from a recorded wav file. But, I only have a SINGLE stream of data (one channel wav.) It appears as if the fastICA won't let me separate more sounds than I have columns in my data matrix. is this correct? If so, is there a work around that is commonly used? Thanks. -- Noah Silverman UCLA Department of

Hi all, I'm searching for a way to exclude outliers from my dataset while making xyplots. While plotting using pairs(), I exclude specific row in my data frame and save the settings as a variable which I later include as an argument: # Discard outliers and save settings as idx idx=with(fieldTrial0809, which(Pro>0 & Pro<0.95 & Fum>0 & Fum<0.4 & Mal>0.1 &

Hello, I am trying to use independent component regression inside the caret package. I ran into the error message below. All entries in the formula are dataframe column names. Any help is appreciated. icrFit <- icr(AUV ~ SPTR + NDDUEAFE + RU20INTR + LBUSTRUU + LF98TRUU + SPTRMDCP,Test,n.comp = 6) Error in fastICA::fastICA(x[, method$ica, drop = FALSE], ...) : unused argument (weights =

Hi, I've got a data frame (556 rows and 36 columns) from which I need to create several xy plots and print to pdf, in order to detect outliers and trends in the data. 16 of the columns contains numerical values, and I would like to create graphs for all combinations. It can be done manually, but creating 256 plots by hand takes time... I guess I have to iterate through the data frame, but

Does anyone know of an ICA (independent components analysis) package for R? In particular I'm looking for an implementation of the FastICA algorithm. Thanks in advance, Cole Harris -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or

Dear list I have a vector of values that allegedly have a bivariate normal distribution. I want to create a plot that shows the values I have obtained, and the bivariate normal distribution curve for the data. Is there a way of doing this in R? Many thanks for your help, Sarah. --------------------------------- [[alternative HTML version deleted]]

Dear Fred, If x1 and x2 are *not* normally distributed, you can use independent component analysis (ICA) which is based on the idea that x will be "more normal" than either x1 and x2 following the central limit theorem. See package(fastICA) by JL Marchini, C Heaton, and BD Ripley for details. HTH Thomas > -----Original Message----- > From: Feng Zhang [mailto:f0z6305 at