similar to: loess' robustness weights in loess

Hi all, I have some difficulties with the lme function and so this is my problem. Supoose i have the following model y_(ijk)=beta_j + e_i + epsilon_(ijk) where beta_j are fixed effects, e_i is a random effect and epsilon_(ijk) is the error. If i want to estimate a such model, i execute >lme(y~vec.J , random~1 |vec .I ) where y is the vector of my data, vec.J is a factor object

Dear R users, I am writing to you because I have a few question on how to fix the error term variances in lme in the hope that you could help me. To my knowledge, the closest possibility is to fix the var-cov structure, but not the whole var-cov matrix. I found an old thread (a few years ago) about this, and it seems that the only alternative is to write the likelihood down and use optim or a

I have a dataset which I want to model using a Poisson distribution, with a given parameter. I would like to know what is the proper way to do a ''goodness of fit'' test using R. I know the steps I''d take if I were to do it ''manually'': grouping the numbers into classes, calculating the expected frequencies using ''ppois'', then

Dear all, Given a LME model (following the notation of Pinheiro and Bates 2000) y_i = X_i*beta + Z_i*b_i + e_i, is it possible to extract the variance-covariance matrix for the estimated beta_i hat and b_i hat from the lme fitted object? The reason for needing this is because I want to have interval prediction on the predicted values (at level = 0:1). The "predict.lme" seems to

Dear All, I'm getting a error while trying to apply the BreastCancer dataset (package=mlbench) to kknn (package=kknn) that I don't understand as I'm new to R. The codes are as follow: rm = (list = ls()) library(mlbench) data(BreastCancer) library(kknn) BCancer = na.omit(BreastCancer) d = dim(BCancer)[1] i1 = seq(1, d, 2) i2 = seq(2, d, 2) t1 = BCancer[i1, ] t2 =

Hello! I would like to use lmer() to fit data, which are some estimates and their standard errors i.e kind of a "meta" analysis. I wonder if weights argument is the right one to use to include uncertainty (standard errors) of "data" into the model. I would like to use lmer(), since I would like to have a "freedom" in modeling, if this is at all possible. For

In a classical meta analysis model y_i = X_i * beta_i + e_i, data {y_i} are assumed to be independent effect sizes. However, I'm encountering the following two scenarios: (1) Each source has multiple effect sizes, thus {y_i} are not fully independent with each other. (2) Each source has multiple effect sizes, each of the effect size from a source can be categorized as one of a factor levels

hi all, Am new to R. I am having problems installing Bioconductor package in R on fedora core 4 running on AMD64 bit machine. this is the error message I get : gcc -shared -L/usr/local/lib -o affyPLM.so avg_log.o biweight.o chipbackground.o common_types.o do_PLMrlm.o do_PLMrma.o do_PLMthreestep.o idealmismatch.o LESN.o lm.o lm_threestep.o log_avg.o matrix_functions.o

Dear sir, I am fitting a glm with default identity link: model<-glm(timetoacceptsecs~maleage*maletub*relweight*malemobtrue*femmobtrue) the model is overdisperesed and plot model shows a low level of linearity of the residuals. The overdispersion and linearity of residulas on the normal Q-Q plot is corrected well by using:

Dear all, my problem is following. I know Stata, but currently I have to use R. Could You please help in finding the analogy to R. (1) creating of City-Dummy. I know in Stata: g byte city=0 replace city=1 if city==12&year==2000 and (2) Create a Time-Dummy-Variable g byte T2000=0 replace T2000=1 if year==2000 (3) I need the City DUmmy for the following combination: I have the

I'm in teaching mode, kernel densities. {History: density() was newly introduced in version 0.15, 19 Dec 1996; most probably by Ross or Robert } When I was telling the students about different kernels (and why their choice is not so important, and "equivalent bandwidths" etc,etc) I wondered about the "Cosine" in my teaching notes which is defined there as k(x)

I am trying to fit a linear mixed model of the form y_ij = X_ij \beta + delta_i + e_ij where e_ij ~N(0,s^2_ij) with s_ij known and delta_i~N(0,tau^2) I looked at the ecme routine in package:pan, but this routine does not allow for different Vi (variance covariance matrix of the e_i vector) matrices for each cluster. Is there an easy way to fit this model in R or should I bite the bullet and

Hi Folks, I've got a question concerning the calculation of the Schwarz-Criterion (BIC) done by summary.regsubsets() of the leaps-package: Using regsubsets() to perform subset-selection I receive an regsubsets object that can be summarized by summary.regsubsets(). After this operation the resulting summary contains a vector of BIC-values representing models of size i=1,...,K. My problem

Dear List, Having had my appetite sufficiently whetted by Prof. Ripley's email about the new graphics capabilities in Unixes, I wanted to try them out. I updated to svn r44608, configured with the following options: R is now configured for x86_64-unknown-linux-gnu Source directory: .. Installation directory: /usr/local C compiler: gcc -O3 -g -std=gnu99

Hello R users! I've been experimenting with lmer to estimate a mixed model with a dichotomous dependent variable. The goal is to fit a hierarchical model in which we compare the effect of individual and city-level variables. I've run up against a conceptual problem that I expect one of you can clear up for me. The question is about random effects in the context of a model fit with a

Hi there, I would like to draw a scatter plot and fit a smooth line by loess. Below is the data. However, the curve line started from 0, which my "resid" list doesn't consist of 0 value. It returned some warnings which I don't know if this is the reason affecting such problem. Here I also attached the warning messages. Please let me know if there is a solution to fix this. Thank

Hi all, I am doing some experiment studies... It seems to me that with different combination of 5 parameters, the end results ultimately converged to two scalars. That's to say, some combinations of the 5 parameters lead to one end result and some other combinations of the 5 parameters lead to the other end result (scalar). I am thinking of this is sort of something like clustering or

I'm updating the loess routines to allow for, among other things, arbitrary local polynomial degree and number of predictors. For now, I've given the updated package its own namespace. The trouble is, panel.loess still calls the original code in package:stats instead of the new loess package, regardless of whether package:loess or package:lattice comes first in the search list. If I

Hi all - I'm stumped by the following mdl <- glm(resp ~ . , data = df, family=binomial, offset = ofst) WORKS yhat <- predict(mdl) WORKS yhat <- predict(mdl,newdata = df) FAILS Error in drop(X[, piv, drop = FALSE] %*% beta[piv]) : subscript out of bounds I've tried without offset, quoting binomial. The offset variable ofst IS in df. Previous postings indicate possible

Hello, I come from a non statistics background, but R is available to me, and I needed to test an implementation of smoothing spline that I have written in c++, so I would like to match the results with R (for my unit tests). I am following Smoothing Splines, D.G. Pollock (available online) where we have a list of points (xi, yi), the yi points are random such that: y_i = f(x_i) + e_i