similar to: problems perfroming the vuong test

Dear All, I am using the function vuong of the package pscl to compare 2 non nested glm models with a numeric response. I did the following m1<-glm(y ~x ,data=xxx) m2<-glm(y ~z , data=xxx) When calling the vuong function I get the following message: > vuong(m1,m2) Error in predprob.glm(m1) : your object of class glm is unsupported by predprob.glmyour object of class lm is unsupported

Dear All, I am using the function vuong from pscl package to compare 2 non nested models NB1 (negative binomial I ) and Zero-inflated model. NB1 <-  glm(, , family = quasipoisson), it is an object of class: "glm" "lm" zinb <- zeroinfl( dist = "negbin") is an object of class: "zeroinfl"   when applying vuong function I get the following: vuong(NB1,

I want to compare the poisson and the zero-inflated poisson distribution on describing the data. So, after using the GLM and the ZEROINFL function, I used the Voung function to compare them. Here is my code: library(pscl) glm1 <- glm(nmer9_1[, 1] ~ 1, family = poisson) zip <- zeroinfl(nmer9_1[, 1] ~ 1) vuong(glm1, zip) However, R returns the following error message: "Error: cannot

Dear All, I would like to know if the Voung test (Voung; Econometrica, 1989) to compare two non-nested regression models has been implemented in R. Thanks in advance for your assistance, mirko [[alternative HTML version deleted]]

Hi, I have two questions. First, I'd like to compare a ZINB model to a negativ binomial model with the Vuong test, but I can't find how to performe it from the zicount package. Does a programm exist to do it ? Second, I'd like to know in which cases we have to use a double hurdle model instead of a zero inflated model. Many thanks, St??phanie Payet REES France R??seau

I would like to test in R what regression fits my data best. My dependent variable is a count, and has a lot of zeros. And I would need some help to determine what model and family to use (poisson or quasipoisson, or zero-inflated poisson regression), and how to test the assumptions. 1) Poisson Regression: as far as I understand, the strong assumption is that dependent variable mean = variance.

Dear All, I fitted two non-nested proportional hazards models using the coxph() function from package survival. Now, I would like to apply a model selection test like, e.g., the likelihood ratio test proposed by Vuong. I found an implementation of Vuong's test in the package 'pscl', but that does not work for cox models. Is there any package available that provides an

Dear all, I want to build a model in R based on animal collection data, that look like the following Nr Village District Site Survey Species Count 1 AX A F Dry B 0 2 AY A V Wet A 5 3 BX B F Wet B 1 4 BY B V Dry B 0 Each data point shows one collection unit in a certain Village, District, Site, and Survey for a certain Species. 'Count' is the number of animals collected in that

Dear UseRs, I have a very simple question. I have a big matrix (say x) including probabilities (values in (0,1)). I have to store in a list the names of the row and the column where max(x) is located. How can I proceed? Thanks in advance for your assistance! mirko

Dear R-users, I am new to the list and I would like to submit (probably!!!!) a stupid question: I found in a paper a reference to a t-test for the evaluationg the difference between the standard deviations of 2 samples. This test is performed in the paper but the methodology is not explained and any reference is reported. Does anyone know where I can find references to this test and if it is

Hello NG, I'm new in this group and I have a problem. the game commandos from pyro studios are not running. I post the error message at the end ... Any hints for me? *********************************** mirko@mirko:~$ wine --debugmsg warn+all /cdrom/Setup.exe warn:dosfs:DRIVE_Init Drive B: not defined warn:dosfs:DRIVE_Init Drive E: not defined warn:dosfs:DRIVE_Init Drive F: not defined

Dear Friends, My problem is related to how to measure probabilities from a probit model by changing one independent variable keeping the others constant. A simple toy example is like this Range for my variables is defined as follows y=0 or 1, x1 = -10 to 10, x2=-40 to 100, x3 = -5 to 5 Model output <- glim(y ~ x1+x2+x3 -1, family=binomial(link="probit")) outcoef <-

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

Hello. I am running 9 poisson regressions with 5 predictors each, using glm with family=gaussian. Gaussian distribution fits better than linear regression on fit indices, and also for theoretical reasons (e.g. the dependent variables are counts, and the distribution is highly positively skewed). I want to determine pseudo R^2 now. However, using the pR2() of the pscl package offers drastically

Hello, I'm running debain on my machine with postfix, squirrelmail and dovecot installed. Everything works fine and I'm able to access my mail via squirrelmail and via Outlook with imap. However I want to create subfolders what is not possible. After reading I discovered that I use mbox and that maildir is needed in order to be able to use subfolders. I have created a Maildir folder with

No, since I'd like to test null: p <= p0 alternative: p > p0. and my understanding is that binom.test tests null: p = p0 (can only be a "simple" null hypothesis according to help(binom.test)) alternative: p > p0 (or p < p0 or p != p0). Thanks, Mirko. > -----Urspr?ngliche Nachricht----- > Von: Douglas Bates [mailto:bates at stat.wisc.edu] >

I'm having some difficulty with graphing outputs of a GLM model I've been working. I have count data for both my predictor (only 1) and response variables, and I have pseudoreplication which I've modeled as a random effect. The odTest() from pscl:: indicated that the negative binomial distribution fit better than Poisson, and I then proceeded by estimating theta from glm.nb. My

Good Day All, I have a negative binomial model which I have developed using the MASS library. I now would like to develop some predictions from it. Running the predict.glm (stats library) using type="response" gives me a non-integer value which was rather puzzling. I would like to confirm that this is actually the mean predicted value of the probability mass function as opposed

FYI -- A small blurb on page 45. -------- Original Message -------- Subject: Shorewall in (IN)SECURE Magazine Date: Tue, 16 Aug 2005 00:10:51 +0200 From: Mirko Zorz <mirko.zorz@net-security.org> To: teastep@shorewall.net Hello Tom, I thought you would be interested to know that Shorewall has been featured in the Software Spotlight section of the third issue of (IN)SECURE, a free security

Hi folks, I was able to track down the problems I had with xemacs / ESS / cygwin / MS W2000 to the file "essd-r.el" in the ESS 5.1.18 package. ESS seems to think, it is running in a complete Unix environment and sends the wrong command line option ("--no-readline" instead of "--ess") to the MS W2000 executable "Rterm.exe". My temporary patch looks