similar to: leaps and bounds

Hi, I used the bic.surv function, S-PLUS functions developed by Chris Volinsky http://www.research.att.com/~volinsky/bma.html, without problems with S-PLUS. I have to use it with R but I am face with a problem: this function call a fortran routine named "leaps" (answer <- .Fortran("leaps", arguments)). I loaded the leaps library, and the leaps function work well with my R,

HI, List I want to know weather any body has used BMA Package for model averaging for prediction of future values.....What are the paprmeters we have to suplly to the model. Any script for the same. thanks in advance ANIL KUMAR( METEOROLOGIST) LRF SECTION NATIONAL CLIMATE CENTER ADGM(RESEARCH) INDIA METEOROLOGICAL DEPARTMENT SHIVIJI NAGAR PUNE-411005 INDIA MOBILE +919422023277

Dear Adrian and Ian (and r-helpers), I encountered a curious result in developing an example using the bicreg function in the BMA package: I noticed that pairs of models with equal R^2 and equal numbers of predictors had nevertheless different BIC values. Looking at the bicreg function, the definition of BIC appears to be the usual one, or close to it [bic <- n * log(1 - r2/100) + (size - 1) *

Hello everybody, In order to predict income for different time points, I fitted a linear model with polynomial effects using BMA (bicreg(...)). It works fine, the results are consistent with what we are looking for. Now, we would like to predict income for a future time point t_next and of course draw the prediction interval around the estimated value for this point t_next. I've found the

Hello everyone, I am searching for a covariate selection procedure in a cox model formulated as a counting process. I use intervals, my formula looks like coxph(Surv(start,stop,status)~ x1+x2+...+cluster(id),robust=T) where id is a country code (I study occurence of civil wars from 1962 to 1997). I'd like something not based on p-values, since they have several flaws for this purpose. I turned

Robust Covariance Estimation Software via Nearest Neighbor Variance Estimation (NNVE) Software to carry out robust covariance estimation by Nearest Neighbor Variance Estimation (NNVE) [Wang and Raftery (2002, J. Amer. Statist. Ass.)] is now available for R and Splus. In the simulation studies published in JASA, this had mean squared error at least 100 times smaller than that of other leading

Robust Covariance Estimation Software via Nearest Neighbor Variance Estimation (NNVE) Software to carry out robust covariance estimation by Nearest Neighbor Variance Estimation (NNVE) [Wang and Raftery (2002, J. Amer. Statist. Ass.)] is now available for R and Splus. In the simulation studies published in JASA, this had mean squared error at least 100 times smaller than that of other leading

The article on BMA (Bayesian model averaging) presents most valuable tools for model selection, but I find one detail confusing in Example 1. In page 4 of RNews 5/2, second paragraph says that the probability of Time variable not being in the model is 0.445. It seems to me that the figure should be 1 - 0.445 = 0.555, because p!=0.445 is the prob. of Time variable being in the model. The plot in

Hi everyone. There's been several threads on baseline hazard in Cox model but I think they were all on cumulative baseline hazard, for instance http://tolstoy.newcastle.edu.au/R/help/01a/0464.html http://tolstoy.newcastle.edu.au/R/help/01a/0436.html "basehaz" in package survival seems to do a cumulative hazard. extract from the basehaz function: sfit <- survfit(fit) H

Dear friends, In R, the help of "bic.glm" tells the difference between postmean(the posterior mean of each coefficient from model averaging) and condpostmean(the posterior mean of each coefficient conditional on the variable being included in the model), But it's still unclear about the results explanations, and the artile of Rnews in 2005 on BMA still don't give more detail on

R version: 0.62.1 (June 15, 1998) I just very naively attempted to grab the 'bicreg' package for Bayesian model selection from the StatLib library, and get it running under R. I've hit a brick wall very quickly, and as an R novice I'm not sure where to go next. Here's what happened: > bicreg(as.matrix(hiv[,c(-17,-18)]),as.matrix(hiv[,18])) Error: invalid formula >

Hi, I am trying to understand the alternative methods that are available for selecting variables in a regression without simply imposing my own bias (having "good judgement"). The methods implimented in leaps and step and stepAIC seem to fall into the general class of stepwise procedures. But these are commonly condemmed for inducing overfitting. In Hastie, Tibshirani and Friedman

Hello R Group, Below is the code I submit: library("BMA") X <- read.table("C:/Documents and Settings/Administrator/Desktop/coding.txt",header=TRUE) Y <- read.table("C:/Documents and Settings/Administrator/Desktop/1DCS.txt",header=TRUE) IGout<- iBMA.glm(X, Y, glm.family= gaussian(), verbose = TRUE, thresProbne0 = 5 ) summary(IGout) IGout I

I may be nuts, but... I would appear that rsync is not setting the exit code properly when it fails. Below are 2 cases where that seems obvious, the line with the '0' after the output is 'echo $?' from bash -- aka the exit code for the last command. Let me know if these are bugs, or I need to somehow get around these situations. Nic >>> starting rsync with

I have been contacted by a researcher who would like to use the bic.logit function (http://lib.stat.cmu.edu/S/bic.logit) for S-PLUS which applies Bayesian Model Averaging to variable selection for logistic regression. I can see that the S-PLUS function uses a call to a Fortran "leaps" function, which does not seem to be available in R. Has this method or a similar method been ported to

Hello I am trying to use all subsets regression on a test dataset consisting of 11 trails and 46 potential predictor variables. I would like to use Mallow's Cp as a selection criterion. The leaps function would provide the required output but does not work with this many variables (see below). The alternative function regsubsets should be used, but I am not able to define the function in

Hi, there, S+ has a function leaps which can be used for criterion based model selection. Is there an equivalent function in R? thanks in advance. Kenny Ye Assistant Professor Department of Applied Math and Statistics SUNY at Stony Brook Stony Brook, New York 11794-3600 (516)632 9344 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read

Dear all I am supposed to use Mallow's Cp creterion to select a model which require a "leaps" package. the version right now I am using is R 1.7.1(os:windows), the "leaps" package is not included. please let me know how and where can I get this package and add it to the current version, any caution on adding this package is highly appreciated. thank you best

Hullo, I am trying to use the leaps package, & keep getting the following error: Error in leaps.setup(x, y, wt = wt, nbest = nbest, nvmax = NCOL(x) + int, : y and x different lengths My data set is attached below. I am rather new to R, & would appreciate any help that could be given. Thanks. http://n4.nabble.com/file/n1755941/lr04.txt lr04.txt -- View this message in context:

Dear UseRs, I wanted to know that i have been using "leaps" for the proper models selection for my work. I read so many articles from internet which categorically outlined that "leaps" command for model selection should never be used as its not efficient. Moreover, as a matter of fact, i personally noticed that the accuracy of "leaps" package is very much