similar to: scatterplot smoothing using gam

All: I have a matrix, X, with a LARGE number of rows. Consider the following three rows of that matrix: 1 1 1 1 2 2 3 3 1 1 1 1 3 3 2 2 3 3 2 2 1 1 1 1 I wish to fit many one-way ANOVAs to some response variable using each row as a set of factors. For example, for each row above I will do something like anova(lm(Y~as.factor(X[1,]))). My problem is that in the above example, I do not want

All: Are there any functions out there for selecting all the points in a region of a plot. I envision something like the identify() function except one could circle a cloud of points (and perhaps a vector would be returned of the same length as the points plotted indicating logical membership in the circled cloud). Perhaps someone has done something with the locator() function that would

All: I am getting an error code from the optimization function. The code is Error in optim(p,fun.LLike, lower=low, upper = up, method = "L-BFGS-B", : non-finite finite-difference value [0] If I add a trace=6 option to my control list the last message before this error is: At X0, 0 variables are exactly at the bounds Any ideas on where I should start would be

running R 1.4.1 on MAC and 1.2.2 on Linux When I use run binom.test with small N the results are a little perplexing to me >binom.test(9,20,p=0.5) gives the below plus other stuff 95 percent confidence interval: 0.2305779 0.6847219 Now: >pbiom(9,20,0.6847219) [1] 0.02499998 # i.e., lower 2.5% of distribution >pbinom(9,20,0.2305779) [1] 0.9923132 >pbinom(8,20,0.2305779)

>Date: Tue, 20 Mar 2001 15:45:52 -0800 >To: Prof Brian D Ripley <ripley at stats.ox.ac.uk> >From: Tony Long <tdlong at uci.edu> >Subject: Re: [R] error starting R 1.2.2 in Redhat 7.0 (GLIB 2.2 not >defined...) >Cc: r-help at stat.math.ethz.c >Bcc: >X-Attachments: > >Brian: > > Thanks!! I figured out how to run "up2date" (that is,

Hello. The laboratory where I work has the possibility to get a Sun machine with 12 processors. My question is: can R be used to implememt parallel algorithms which take into account the several processors? Is there a way to run *threads* in R? If this is possible, would somebody share his/her experiences? Thanks. Isabel -- Isabel Cañette Iguá 4225 Centro de Matemática 11400

All: I just upgraded my system to Redhat 7.0, and installed the RPM for R 1.2.2. I intall R 1.2.2 using the rpm -U xxxx command. When R is invoked from some directories it starts and runs fine, from other directories I get a strange message at startup (and R exits without starting). Thus far the directory with the problem tends to contain ".data" files from the previous version

All: I am attempting to allow multiple users on a linux system use some common functions. The problem I am having is that for some users their default library location is .lib.loc = "/usr/lib/R/library" whereas for other users it is .lib.loc = "/usr/local/lib/R/library" It seems as though the packages (for example cluster) load fine if I just change .lib.loc to

Perhaps this is obvious...but is there anyway to create a matrix like object with has more than 2 dimensions. I want to estimate a bunch of var/cov matrices inside an index loop. It would be more computationally efficient to do this in C...but R makes many of the operations inside the loop really easy! Thanks. -- Tony Long Ecology and Evolutionary Biology Steinhaus Hall University of

All: I apologize as I think I had this problem before, and misplaced my notes on the solution. I just tried to upgrade my version of R on a Redhat Linux computer. I downloaded and ran the rpm for R-recommended-1.4.1-1.i386.rpm Problem is that when I type R from any directory the "old" version still seems to run. I can only guess this has to do with where the new and old

I am trying to use dynamic loading of an outside C routine. I am attempting 6.12.1 of Phil Spector's book. When I try to load the object file I get an error I don't understand: > dyn.load("runa.o") Error in dyn.load(x) : unable to load shared library "/usr/home/tdlong/run_avg/runa.o": /usr/home/tdlong/run_avg/runa.o: ELF file's phentsize not the expected

Sorry: I should have reproduced the "problem" with a simple example. I do this below. I think there is likely a switch I can change using par, but don't know what it is. The problem is the tick marks for the Y- axis are only on plots in column #1 and for the X-axis in row # 2. Tony x <- 1:10 y <- 1:10*5 par(mfrow=c(2,2)) plot(x,y) plot(x,y) plot(x,y) plot(x,y)

I am running R version ??? under Redhat 5.2. It seems as though the --nsize object has no effct on the size of the allocated Ncells as determined using gc(). Yes, I have that much data.... That is if I envoke R with R --vsize 100 --nsize 5000000 then type gc() I get free total Ncells 92202 200000 Vcells 12928414 13107200 Thanks Tony Long Ecology and Evolutionary Biology Steinhaus

Dear R Users, Before running ksmooth( ), a suitable bandwidth selection is needed. I use some functions for this task and receive these results for my data: width.SJ(y,nb=100,method="ste") : 40.25 bcv(y,nb=100) : 40.53 ucv(y) : 41.26 bandwidth.nrd(y) : 45.43 After implementing the function ksmooth(x,y, bandwidth= each of abovementioned bandwidths), I have some NAs

Apologies in advance if my comments don't help, in which case, no need to respond, but I noted in ?ksmooth: "bandwidth the bandwidth. The kernels are scaled so that their quartiles (viewed as probability densities) are at ? 0.25*bandwidth." So, could this be a source of the discrepancies you cited? Given that ?ksmooth explicitly says: "Note: This function was implemented for

Hello all, I would like to use the Nadaraya-Watson estimator assuming a Gaussian kernel: So far I sued the library(sm) library(sm) x<-runif(5000) y<-rnorm(5000) plot(x,y,col='black') h1<-h.select(x,y,method='aicc') lines(ksmooth(x,y,bandwidth=h1)) which works fine. What if my data were clustered requiring a bandwidth that varies with x? How can I do that? Thanks in

Dear Sir, Madam, or to whom this may concern, my name is Jan Failenschmid and I am a Ph.D. student at Tilburg University. For my project I have been looking into different types of kernel regression estimators and corresponding R functions. While comparing different functions I noticed that stats::ksmooth returned different estimates for the same bandwidth as other kernel regression estimators

I am working with a model that I have to estimate a nonparametric function. The model is partial linear i.e. Y=X$\beta$ + f(z) + $\epsilon$ I am using the ' double residual methods' Robinson (1988) Speckman (1988) where I estimate a nonparametric function for each of the parametric variables in terms of the nonparametric one i.e. X[,i]=g(Z)+ u this is done because I need the $E(

I'd like to superimpose a ksmooth() onto a barplot(). My data is: > d 2009-06-20 2009-06-21 2009-06-22 2009-06-23 2009-06-24 2009-06-25 2009-06-26 2009-06-27 2009-06-28 2009-06-29 2009-06-30 2009-07-01 2009-07-02 Same Breed (B) 12.64 21.08 13.52 12.51 13.71 9.91 14.24 7.18 11.81 5.92 12.04 17.96

All: Any ideas on how to do variance component estimation, or at least get Expected Mean Sqaures out of the modeling commands. Thanks. Tony Tony Long Ecology and Evolutionary Biology Steinhaus Hall University of California at Irvine Irvine, CA 92697-2525 Tel: (949) 824-2562 (office) ****NOTE NEW AREA CODE**** Tel: (949) 824-5994 (lab) ****NOTE NEW AREA CODE****