similar to: glm families in R

Two questions: 1. inverse.gaussian is up there as one of the glm families, but do people ever use it? There is no inverse.gaussian in the R distribution family, and when I checked McCullagh & Nelder, it only appeared twice in the book (according to subject index), once in the table on p. 30 and once on p. 38 in a passing sentence. Is there a good reference on this distribution? 2. When I

Any thoughts on the following? Am I missing something? [I am running R for Windows 0.63.1 on a Win NT 4.0 (workstation, sevice pack #4) box.] Thanks. FC. R : Copyright 1998, The R Development Core Team Version 0.63.1 Beta (Dec 5, 1998) R is free software and comes with ABSOLUTELY NO WARRANTY. You are welcome to redistribute it under certain conditions. Type "?license" or

FWIW p. 206 of the White Book gives the following for names(binomial()): family, names, link, inverse, deriv, initialize, variance, deviance, weight. So $aic wasn't there In The Beginning. I haven't done any more archaeology to try to figure out when/by whom it was first introduced ... Section 6.3.3, on extending families, doesn't give any other relevant info. A patch for

Is it generally known/has it been previously discussed here that the $aic() component in GLM-family objects (e.g. results of binomial(), poisson(), etc.) does not as implemented actually return the AIC, but rather -2*log-likelihood + 2*(model_has_scale_parameter) ? Can anyone in this forum gauge how a documentation patch would be received? This behaviour does not seem to be documented in ?family

Hi all I am converting an S-Plus function into R. The S-Plus code uses some of the glm families, and family objects. The family objects in S-Plus and R have many different features, for example: In R: > names(Gamma()) [1] "family" "link" "linkfun" "linkinv" "variance" [6] "dev.resids" "aic"

I have just upgraded from Red Hat Linux 5.2 to version 6.0 (with GNOME) on an intel box. I have also upgraded to the latest R version (0.64.1-2 rpm) and have also installed the corresponding GNOME rpm. However, I am not able to start R: [cribari at edgeworth install]$ R /usr/lib/R/bin/R.binary: error in loading shared libraries: /usr/lib/R/bin/R.binary: undefined symbol: __setfpucw [cribari at

When I try to install the R RPM under Mandrake Linux 7.1 (intel), I get the following error: [root at edgeworth programs]# rpm -ivh R-base-1.1.1-1.i386.rpm error: failed dependencies: libreadline.so.3 is needed by R-base-1.1.1-1 [root at edgeworth programs]# I do have realine installed: [root at edgeworth programs]# rpm -qa | grep readline readline-4.1-4mdk and

Does anyone have an R function for the estimation of Tobit regression models (for censored responses), as described, e.g., in Davidson, R. and MacKinnon, J.G., 1993, Section 15.7, Estimation and Inference in Econometrics, Oxford University Press? Thanks. FC. -- Francisco Cribari-Neto voice: +55-81-32718420 Departamento de Estatistica fax: +55-81-32747425 Universidade

I am trying to install R under Red Hat Linux 5.2, but the following error happens: [root at edgeworth cribari]# rpm -ivh R-base-0.63.2-1.i386.rpm failed dependencies: libncurses.so.3.0 is needed by R-base-0.63.2-1 I must be missing something very simple. I would appreciate any guidance. Thanks. FC. -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help

I am running R 1.2.1 under Mandrake Linux 7.2 (installed from RPM) and I installed a few packages using install.packages() . Now, I am trying to upgrade these packages using update.packages() but I believe I am running into some kind of error. See below. Any tips will be greatly welcome. Thanks in advance. Francisco PS: update.packages() only tries to update wle from version 0.3 to 0.4, gives

Dear R-Users, I wish to create a new family object based on the Binomial family. The only difference will be with the link function. Thus instead if using the 'logit(u)' link function, i plan to use '-log(i-u)'. So far, i have tried to write the function following that of the Binomial and Negative Binomial families. The major problem i have here is with the definition of the

Hi Folks, I am interested in extending the repertoire of link functions in glm(Y~X, family=binomial(link=...)) to include a "tan" link: eta = (4/pi)*tan(mu) i.e. this link bears the same relation to the Cauchy distribution as the probit link bears to the Gaussian. I'm interested in sage advice about this from people who know their way aroung glm. >From the surface, it looks

Hello, I'd be very grateful for help with the following. I attempt to run a basic GLM as follows: basic_model=glm(factor(A$b) ~ factor(A$c) ) variables b and c are text values and their class is factor. they have no missing values. I get the error: Error: NA/NaN/Inf in foreign function call (arg4) In addition: warning messages: 1: Ops.factor(y,mu) : - not meaningful for factors 2:

G'day all, I had a look at the GLM code of R (1.4.1) and I believe that there are problems with the function "glm.fit" that may bite in rare circumstances. Note, I have no data set with which I ran into trouble. This report is solely based on having a look at the code. Below I append a listing of the glm.fit function as produced by my system. I have added line numbers so that I

I am running R 1.2.2 under Linux. When using the function stl in the ts library, how can I save the seasonal component? What I would like was something like: library(ts) data(nottem) data.stl <- stl(nottem, "per") x <- data.stl$sea This what I get: > x NULL I would, however, like to store in x the seasonal component. Thanks in advance. Francisco. -- Francisco

Hi R-users,   I wrote a code to evaluate double sum as follows:   ff2 <- function(bb,eta,z,k) { r <- length(z) for (i in 1:r) { sm1 <- sum((z[i]*bb/2)*(psigamma((0:k)+eta+1,deriv=0)/(factorial(0:k)*gamma((0:k)+eta+1))))  sm2 <- sum((besselI(z[i]*bb,eta)*log(z[i]*bb/2) - sm1)/besselI(z[i]*bb,eta))  sm2 } ff2(bb,eta,z,10)     but it gave me the following message:   >

In R, family has aic component since version 0.62. There is no aic component in family in R 0.61.3. Looking at blame, https://github.com/wch/r-source/blame/tags/R-0-62/src/library/base/R/family.R , aic component in family is introduced in svn revision 640 (https://github.com/wch/r-source/commit/ac666741679b50bb1dfb5ce631717b375119f6ab): using aic(.) [Jim Lindsey]; use switch() rather than many if

Dear all, I am using the book "Generalized Linera Models and Extension" by Hardin and Hilbe (second edition, 2007) at the moment. The authors suggest that instead of OLS models, "the log link is generally used for response data that take only positive values on the continuous scale". Of course they also suggest residual plots to check whether a "normal" linera model

Let me first tell you that you and your cohorts are doing a great service to the statistical community with R. I have been working with Rseptbeta under Win95 and I am very impressed with its capabilities. I thought I give you some feedback gained from my limited playing with R. Fritz Scholz fritz.scholz at boeing.com ============================================================= When

Dear list: I am attempting to use what I thought would be a pretty straightforward practical application of Cox regression. I figure users of the survival package must have come across this problem before, so I would like to ask you how you dealt with it. I have set up an illustrative example and included it at the end of this post. I took a sample of 100 data points from each of two populations