similar to: Comparison of glm.nb and negbin from the package aod

Hi, When modeling with negbin from the aod package, parameters for a given count y | lambda~Poisson(lambda) with lambda following a Gamma distribution Gamma(r, theta) are estimated. The intercept is called phi. Some other parameters may be also be estimated from factors in the data: the estimates returned for all these would be in accordance with the Value listing in the negbin entry in the aod

I have tried glm.nb in the MASS package, but many models (I have 250 models with different combinations of predictors for fish counts data) either fail to converge or even diverge. I'm attempting to use the negbin function in the AOD package, but am unsure what to use for the "random" term, which is supposed to provide a right hand formula for the overdispersion parameter.

Hi, I wonder if predict.gam is supposed to work with family=negbin() definition? It seems to me that the values returned by type="response" are far off the observed values. Here is an example output from the negbin examples: > set.seed(3) > n<-400 > dat<-gamSim(1,n=n) > g<-exp(dat$f/5) > dat$y<-rnbinom(g,size=3,mu=g) >

Dear R people, I am not sure if this is the correct place to tell about problems in evolving programmes, but it seems that the `summary.negbin' function of the excellent `MASS' library is now broken, and gives the following error message: > summary(hm) Error in summary.negbin(hm) : subscript out of bounds `summary.negbin' calls `summary.glm' which seems to work and give the

Dear R-collegues, I'm getting an error message (Error in round) when summarising a glm.nb model, and when using anova.negbin (in R 1.3.1 for windows): > m.nb <- glm.nb(tax ~ areal) > m.bn Call: glm.nb(formula = tax ~ areal, init.theta = 5.08829537115498, link = log) Coefficients: (Intercept) areal 3.03146 0.03182 Degrees of Freedom: 283 Total (i.e. Null); 282

Dear Clara, Thanks for the reply. I am forwarding your message to the list, ok. When I wrote was a way of get further information to help the helpers. happy holidays, milton ---------- Forwarded message ---------- From: Clara Brück <clara_brueck@web.de> Date: 2009/12/30 Subject: Re: [R] Negbin Error Warnings To: milton ruser <milton.ruser@gmail.com> Dear Milton, Thanks for

Hi, I ran a negative binomial regression (NBR) using the Zelig-package and the negbin model. When I then try to use the simumlation approach using the setx () and sim() functions to calculate expected values and first difference for different levels of one of my independent variables, I get 50 errors warnings, telling me that the calculation rpois produced NAs. However, the data I use

Information on package 'aod' Description: Package: aod Version: 1.1-2 Date: 2005-06-08 Title: Analysis of Overdispersed Data Author: Matthieu Lesnoff <matthieu.lesnoff at cirad.fr> and Renaud Lancelot <renaud.lancelot at cirad.fr> Maintainer: Renaud Lancelot <renaud.lancelot at cirad.fr> Depends: R (>=

Information on package 'aod' Description: Package: aod Version: 1.1-2 Date: 2005-06-08 Title: Analysis of Overdispersed Data Author: Matthieu Lesnoff <matthieu.lesnoff at cirad.fr> and Renaud Lancelot <renaud.lancelot at cirad.fr> Maintainer: Renaud Lancelot <renaud.lancelot at cirad.fr> Depends: R (>=

Edward, you also can use the package aod on CRAN, see the help page of the function negbin. Best Matthieu An example: > library(aod) > data(dja) > negbin(y ~ group + offset(log(trisk)), ~group, dja, fixpar = list(4, 0)) Negative-binomial model ----------------------- negbin(formula = y ~ group + offset(log(trisk)), random = ~group, data = dja, fixpar = list(4, 0))

Dear R users, I am trying to use "gamm" from package "mgcv" to model results from a mesocosm experiment. My model is of type M1 <- gamm(Resp ~ s(Day, k=8) + s(Day, by=C, k=8) + Flow + offset(LogVol), data=MyResp, correlation = corAR1(form= ~ Day|Mesocosm), family=poisson(link=log)) where the response variable is counts, offset by the

I am getting the following erro rmessage in ordistep. I have a number of similarly structured datasets using ordistep in a loop, and the message only occurs for some of the datasets. I cannot include a reproducible sample - the specific datasets where this is occur ing are fairly large and there are several pcnm's in the rhs of the formula. thanks for any pointers that may allow me to

Folks, Thanks for the help with the hier.part analysis. All the problems stemmed from an import problem which was solved with file.chose(). Now that I have the variables that I'd like to use I need to run some GLM models. I think I have that part under control but I'd like to use a jackknife approach to model validation (I was using a hold out sample but this seems to have fallen out

Dear R crew: I am sorry this question has been posted before, but I can't seem to solve this problem yet. I have a simple dataset consisting of two variables: cestode intensity and chick size (defined as CAPI). Intensity is a count and clearly overdispersed, with way too many zeroes. I'm interested in looking at the association between these two variables, i.e. how well does chick

I am using gam in mgcv (1.3-22) and trying to use gcv to help with model selection. However, I'm a little confused by the process of assessing GCV scores based on their magnitude (or on relative changes in magnitude). Differences in GCV scores often seem "obvious" with my poisson gams but with negative binomial, the decision seems less clear. My data represent a similar pattern as

How might you fit a generalized linear model (glm) with variance = mu+theta*mu^2 (where mu = mean of the exponential family random variable and theta is a parameter to be estimated)? This appears in Table 2.7 of Fahrmeir and Tutz (2001) Multivariate Statisticial Modeling Based on Generalized Linear Models, 2nd ed. (Springer, p. 60), where they compare "log-linear model fits to

Dear R crew: I think I am in the right mailing list. I have a very simple dataset consisting of two variables: cestode intensity and chick size (defined as CAPI). Intensity is clearly overdispersed, with way too many zeroes. I'm interested in looking at the association between these two variables, i.e. how well does chick size predict tape intensity? I fit a zero inflated negat. binomial

Hello, I am having some difficulty converting my gam code to a correct gamm code, and I'm really hoping someone will be able to help me. I was previously using this script for my overdispersed gam data: M30 <-gam(efuscus~s(mic, k=7) +temp +s(date)+s(For3k, k=7) + pressure+ humidity, family=negbin(c(1,10)), data=efuscus) My gam.check gave me the attached result. In order to

Hi there, I'm using presence-absence data in a gam (i.e. 0 or 1 as values) I am trying to run a gam with 'dummy covariates' i.e. 1~1 unfortunately my model: * model<-gam(1~1, data=bats, family=negbin)* keeps putting out: * Error in gam(1 ~ 1, data = bats, family = negbin) : Not enough (non-NA) data to do anything meaningful* Is there a specific reason it would do this? I have

Hi, I am new to R and has been depending mostly on the online tutotials to learn R. I have to deal with zero inflated negative binomial distribution. I am however unable to understand the following example from this link http://www.ats.ucla.edu/stat/r/dae/zinbreg.htm The result gives two blocks. *library(pscl) zinb<-zeroinfl(count ~ child + camper | persons, dist = "negbin", EM =