similar to: Deviance of zeroinfl/hurdle models

Dear list, Can anyone tell me how to obtain the rank of a sparse Matrix, for example from package Matrix (class dgCMatrix)? Here is an example of QR decomposition of a sparse matrix (from the sparseQR class help). library(Matrix) data(KNex) mm <- KNex$mm str(mmQR <- qr(mm)) Similarly, using the functions/classes from the relatively new MatrixModels package: library(MatrixModels)

Dear list, I am trying to plot a radial dendrogram using the ape package, which requires my data to be of class 'phylo'. Currently I have my dendrogram stored as an object of class 'dendrogram' which was produced from an outside bit of C code, but was made into an object of class 'igraph.eigenc' and converted to a dendrogram using 'as.dendrogram()' from the igraph

Dear list, I have been comparing the outputs of two packages for latent class regression, namely 'flexmix', and 'mmlcr'. What I have noticed is that the flexmix package appears to come up with a much better fit than the mmlcr package (based on logLik, AIC, BIC, and visual inspection). Has anyone else observed such behaviour? Has anyone else been successful in using the mmlcr

Hello list, Has anyone had any luck creating an M-step driver for negative binomial regression for use with package flexmix? I've had a look here: http://cran.r-project.org/web/packages/flexmix/vignettes/flexmix-intro.pdf as well as poking around in the flexmix source, but I haven't had much luck getting anything to work. I can't figure out how to a) come up with an initial estimate

Dear list, I posted essentially this same question to the r-sig-geo mailing list last week with no response :( Unfortunately I am no closer to reaching a solution, so I now post it here (with some clarifications) in the hope that someone following this list might have an answer for me: Has anyone had much experience with spatial interaction (or gravity) models, specifically in the form of

Dear list, I have been working with the MatrixModels package quite a bit this week, and it is proving to be extremely valuable for my current work (I am working with several factors with many levels, leading to a sparse model matrix). However, as my knowledge of statistical theory leaves much to be desired, there are certain aspects of model evaluation etc that I am having trouble with. Has

Hello list, I am not sure if the terminology that I am using here is widely used, however, I provide an example in the hopes that my problem will become clear. My basic problem is that I am unsure of how to 'constrain' my model estimates to reproduce the aggregate (by factor levels) observed dependent variable for a negative.binomial model. I realize this sounds rather vague, so I provide

Hi, I am wondering if there is an easy way to access the hat matrix for zeroinfl and hurdle objects in the pscl library? Thanks, Chris [[alternative HTML version deleted]]

Hello, I have a data frame with 1425 observations, 539 of which are zeros. I am trying to fit the following ZINB: f3<-formula(Nbr_Abs~ Zone * Year + Source) ZINB2<-zeroinfl(f3, dist="negbin", link= "logit", data=TheData, offset=log(trans.area), trace=TRUE) Zone is a factor with 4 levels, Year a factor with 27 levels, and Source a factor with 3 levels. Nbr_Abs is counts

Good afternoon, I?m in need of an advice regarding a proper use of glm.nb, zeroinfl or hurdle with my dataframe. I can not provide a self-contained example, since I need an advice on this current dataset and its ?contradictory? results. So.... i have a dataset which contains 1309 cases and 11 variables, highly right-skewed and heavily zeroinflated (with over 1100 cases that have 0 value

Dear all, A question related to the following has been asked on R-help before, but I could not find any answer to it. Input will be much appreciated. I got an unexpected sign of the "slope" parameter associated with a covariate (diam) using zeroinfl(). It led me to compare the estimates given by zeroinfl() and hurdle(): The (significant) negative estimate here is surprising, given

Hi all, I have been trying to use zeroinfl() with the pscl package with R version 2.1.1. and with the newest versions of the contrib packages compatible with R 2.1.1. I have read the examples, the vignette and all the posts relating to zeroinfl() but I am still confused as to how to structure the model. Here is a small example; the error message is the same for big data sets

I'm trying to fit a zero-inflated poisson model using zeroinfl() from the pscl library. It works fine for most models I try, but when I include either of 2 covariates, I get an error. When I include "PopulationDensity", I get this error: Error in solve.default (as.matrix(fit$hessian)) : system is computationally singular: reciprocal condition number = 1.91306e-34 When I

Hi Just a quick problem that I hope is simple to resolve. I'm doing some work with zero inflated poisson models using the pscl package. I can build models using zeroinfl and get outputs fom them with no problem, but when I try to use the predict.zeroinfl function, I get Error: could not find function "predict.zeroinfl". I was using an older version of R, but still had the same

Dear list, I am running some zero inflated models and would like to know what the deviance of the models. Unlike running a normal GLM where the deviance is displayed in the summary all that is displayed in a summary of the zero inflated model is the log likelihood. I hope this isn't a read the manual question, and if it is I apologize for wasting your time, but if you could still send me a

Hello all: I have a question regarding the fitted.values returned from the zeroinfl() function. The values seem to be nearly identical to those fitted.values returned by the ordinary glm(). Why is this, shouldn't they be more "zero-inflated"? I construct a zero-inflated series of counts, called Y, like so: b= as.vector(c(1.5, -2)) g= as.vector(c(-3, 1)) x <- runif(100) # x

Dear R-Helpers, Can anyone please help me to interpret warning messages from zeroinfl (package pscl) while fitting a zero inflated negative binomial model? The console reports convergence and the parameters seam reasonable, but these <<Warning messages: 1: algorithm did not converge in: glm.fit(X, Y, family = poisson()) 2: fitted rates numerically 0 occurred in: glm.fit(X, Y, family =

Hi, I am a little confused at the output from predict() for a zeroinfl object. Here's my confusion: ## From zeroinfl package fm_zinb2 <- zeroinfl(art ~ . | ., data = bioChemists, dist = "negbin") ## The raw zero-inflated overdispersed data > table(bioChemists$art) 0 1 2 3 4 5 6 7 8 9 10 11 12 16 19 275 246 178 84 67 27 17 12 1 2 1 1

I am attempting to fit a full zero-inflated Poisson model then use backward elimination to arrive at the best-fitting model. When I try to fit the model with zeroinfl() I get this error: Error in while (abs((ll_old - ll_new)/ll_old) > control$reltol) { : missing value where TRUE/FALSE needed Any suggestions? Thanks for your help! Paige Barlow MS Student Virginia Tech Dept Fish

I am fitting a model using zeroinfl() and it runs without errors, returning results that are generally consistent with my hypotheses. One of my variables is percent black (pblack). This variable was highly significant in some of the other count models I ran on the way to my current formulation. It is not significant in this model. As such I decided to try adding pblack^2 to the model to see