search for: raftery

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 covariance estimators when the proportion of outliers was high. cov.nnve is now available in the covRobus...

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 covariance estimators when the proportion of outliers was high. cov.nnve is now available in the covRobus...

Forwarded from Adrian Raftery: -----Original Message----- From: Adrian Raftery [mailto:raftery at stat.washington.edu] Sent: Tuesday, October 22, 2002 10:15 AM To: Warnes, Gregory R Subject: Re: FW: [R] Mixture Transition Distribution (MTD) time series model in R or S ? Thanks, Greg. Current MTD software is reviewed in Andr?...

...does not correspond to the number of observations for Poisson regression/log-linear models, i.e. when family="poisson" or family="quasipoisson". This sounds dangerous since nobs() is, as the documentation states, primarily aimed at computing the Bayesian information criterion. Raftery (1995:20) warned against this: > What should n be? Once again, it is best to use the actual number of > individuals, i.e. the sum of the cell counts, and not the number of > cells (Raftery, 1986a). Is there a reason why this should not/cannot be done that way? This behavior can be reprod...

Package: mcgibbsit Title: Warnes and Raftery's MCGibbsit MCMC diagnostic Version: 1.0 Author: Gregory R. Warnes <gregory_r_warnes at groton.pfizer.com> Description: mcgibbsit provides an implementation of Warnes & Raftery's MCGibbsit run-length diagnostic for a set of (not-necessarily independent) MCMC sampers. It c...

Package: mcgibbsit Title: Warnes and Raftery's MCGibbsit MCMC diagnostic Version: 1.0 Author: Gregory R. Warnes <gregory_r_warnes at groton.pfizer.com> Description: mcgibbsit provides an implementation of Warnes & Raftery's MCGibbsit run-length diagnostic for a set of (not-necessarily independent) MCMC sampers. It c...

Hello all (especially MCLUS users). I'm trying to make use of the MCLUST package by C. Fraley and A. Raftery. My problem is trying to figure out how the (model) identifier (e.g, EII, VII, VVI, etc.) relates to the covariance matrix. The parameterization of the covariance matrix makes use of the method of decomposition in Banfield and Rraftery (1993) and Fraley and Raftery (2002) where Sigma_k = lambda_k...

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

...| Albert Einstein ------------------------------------------------------------------------------- ---------- Forwarded message ---------- Date: Tue, 9 Feb 1999 10:06:57 -0800 (PST) From: Richard Fairfield <rcf at ms.washington.edu> To: "Adrian E. Raftery" <raftery at stat.washington.edu> Cc: Statistics Help <help at stat.washington.edu> Subject: Re: installing R on monto Adrian: We are unable to compile this software on either platform. I don't think there is anything more we can do at this time. Here are several options f...

...me observations are only a few days long. With n at the denominator (Letué's criteria), the penalized term would be so low that it's like not having it: >log(7000)/7000 [1] 0.001264809 (where loglik from summary(cox.fit) range from -155 to -175, dependig on the model) b) Volinsky & Raftery "propose a revision of the penalty term in BIC so that the penalty is defined in terms of the number of uncensored events instead of the number of observations." (Volinsky & Raftery , Bayesian Information Criterion for Censored Survival Models, June 16, 1999, http://www.research.att...

...aset supplied with R using boxplot(iris$Sepal.Length ~ iris$Species). You see that the boxplot is the same before and after the fix. Does anybody know why this occurs, and how I can get around this issue? Thanks, -- Regards, Paul ===================== Contact Details ===================== Paul Raftery, BEng(Hons) (Mech), Fulbright Fellow, PhD http://www.paulraftery.com/ <http://www.paulraftery.com/> Postdoctoral Research Engineer Informatics Research Unit for Sustainable Engineering (IRUSE) http://www.iruse.ie/ Department of Civil Engineering, National University of Ireland, Galway, Univ...

Hi all I have a couple of obscure questions for R/Splus experts (which unfortunately isn't me!) I am trying to compute Bayes Factors using some Splus code of Raftery in Gilks et al (1996). Only problem is 1) R doesn't seem to have a robust covariance (cov.mve) which I suspect I need rather than a non-robust classical estimate 2) Splus has cov.mve BUT dgamma in Splus doesn't have a scale parameter (like in R). I presume it is easy to include one but it...

I was sniffing in the fracdiff library (this is for fractionally integrated ARMA processes; Haslett and Raftery 1989). The documentation suggests that one tries the following simple example: library(fracdiff) ts.test <- fracdiff.sim( 5000, ar = .2, ma = -.4, d = .3) fracdiff( ts.test$series, nar = length(ts.test$ar), nma = length(ts.test$ma)) When I run this, I get the following error: R --vanilla &lt...

It is going to be easy question to you. I've started to interest in model-based clustering. Adrian E. Raftery "Recent Advances in Model-Based Clustering: Image Segmentation and Variable Selection" (www.stat.washington.edu/Raftery)showed that we can compare different classification methods using BIC statistic. For diabetes dataset the best model is VVV model with 3 classes- for this model the BIC...

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. On the Bayesian averaging homepage http://www.research.att.com/~volinsky/bma.html I found some S code some of which perhaps may run in R. There was a call to an algorithm possibly within S but not supported by R 64.1: "leaps and bounds". I guess it is a minimization step. Can anyone clarify the algorithm and perhaps even give a pointer to some code ? I guess this may be

...and Rousseeuw, P.J. (1990). Finding Groups in Data: An Introduction to Cluster Analysis. Wiley, New York. Struyf, A., Hubert, M. and Rousseeuw, P.J. (1997). Integrating Robust Clustering Techniques in S-PLUS, Computational Statistics and Data Analysis, 26, 17-37. (mclust) D. Banfield and A. E. Raftery, Model-based Gaussian and non-Gaussian Clustering, Biometrics, 49:803-821 (September 1993). C. Fraley, Algorithms for Model-based Gaussian Hierarchical Clustering, Technical Report No. 311, Department of Statistics, University of Washington (October 1996), to appear in SIAM Journal on Scientific...

Hi everyone, I am wondering if there exists a stepwise regression function for the Bayesian regression model. I tried googling, but I couldn't find anything. I know "step" function exists for regular stepwise regression, but nothing for Bayes. Thanks -- View this message in context: http://www.nabble.com/Bayesian-regression-stepwise-function--tp26013725p26013725.html Sent from

Thanks for telling me that you could not get my message, I hope this work better... so my question was: I built a population matrix to which I applied the fonction eigen in order to find the main parameters about my population. I know that the first eigen value correspond to lambda or exponential growth rate of my population. My problem is that I want to have the 95% confidence interval of the

Dear r-users, I am looking for an implementation of the Naive Bayes classifier for a multi-class classification problem. I can not even find the Naive Bayes classifier for two classes, though I can not believe it is not available. Can anyone help me? Uschi -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read