similar to: Viewing code

Hi, Currently the `gaussian' family's initialization code signals an error if any response data are zero or negative and a log link is used. Given that zero or negative response data are perfectly legitimate under the GLM fitted using `gaussian("log")', this seems a bit unsatisfactory. Might it be worth changing it? The current offending code from `gaussian' is:

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

Hi everyone: I am just wondering how can I use weights with MCMClogit function (in MCMCpack package). For example, in case of glm function as given below, there is weights option in the arguments. Aparently there is no option of using weights in MCMClogit. glm(formula, family = gaussian, data, weights, subset, na.action, start = NULL, etastart, mustart, offset, control =

Hello, Suppose one is interested in fitting a GLM with a log link to binomial data. How does R choose starting values for the estimation procedure? Assuming I don't supply them. Thanks, Tyler

Dear all, we are trying to model some data from rare plants so we always have less than 50 1x1 km presences, and the total area is about 550.000 square km. So we have a real problem, when we perform a GAM, if we consider only the same amount of absences than presences. We have thought to use a greater number of absences but in this case we shoud downweight them. Does anybody know how to use the

Hello R Users, I am having difficulty deleting the last warning message in a loop so that the only warning that is produced is that from the most recent line of code. I have tried options(warn=1), rm(last.warning), and resetting the last.warning using something like: > warning("Resetting warning message") This problem has been addressed in a previous listserve string,

I encountered a segfault running glm() and wonder if it could have something to do with the way memory is handled in a dual core system (which I just set up). I'm running R-base-2.4.0-1, installed from the SuSE 10.1 x86_64 rpm (obtained from CRAN). (My processor is an AMD Athlon 64 x2 4800+). The error and traceback are *** caught segfault *** address 0x8001326f2b, cause 'memory not

Dear List, I'm fairly new in R. I'd like to see how glm() uses the argument "family" in fitting a model. Specifically, I'd like to see how a glm with a gamma family is fitted. Thanks for any help, Axel. [[alternative HTML version deleted]]

In the following code, the only difference between the two plots is the order the variables are plotted. In this case, the plot of "cdata.den" in plot #1 is different from its plot in #2. Specifically, "cdata.den" spans the x-axis from -5 to 30 in plot #1 and from 0 to 20 in plot #2. Does anyone understand why these two plots do not yield the same result? #Make density

Bug summary: glm() causes a segfault if the argument 'data' is a data frame with more than 16384 rows. Bug demonstration: -------input --------------- N <- 16400 df <- data.frame(x=runif(N, min=1,max=2),y=rpois(N, 2)) glm(y ~ x, family=poisson, data=df) ------ output --------------- *** caught segfault *** address (nil),

I am simulating sickness among a group of families. Part of the task is to randomly draw who in the family will be sick, randomly drawing from family ID's where Dad =1, Mom = 2, Kid1 = 3, Kid2 = 4., etc. My census of Dads is of the form shown below. Dad_ID Spouse (Y=1;N=0) #Kids #People_Becoming_Sick 1 1 0 1 2

I'm fitting some small data sets using a binomial glm with complementary log-log link. In some ill-conditioned cases I am getting convergence failures. I know how to adjust maxit and epsilon, but that doesn't seem to help. In fact I know some very good starting values for my fits, but I can't see how to get them in to glm(). How may I do this? -- Dr Murray Jorgensen

All - I am evaluating an arima(2,1,3) and arima(3,1,3) and notice the log-likelihood of the restricted model is higher than the log-likelihood of the unrestricted. Since these are nested models, I thought the unrestricted model would have a log-likelihood at least as large as that of the restricted model. Am I interpreting the "loglik" output incorrectly? Regards, Stephen

Hi there I want to pass arguments (i.e. the response variable and the subset argument) in a self-made function to glm. Here is one way I can do this: f.myglm <- function(y,subfact,subval) { glm(d.mydata[,y]~d.mydata[,'x1'],family=binomial,subset=d.mydata[,subfact]==subval) } > str(d.mydata) `data.frame': 15806 obs. of 3 variables: $ y : Factor w/ 2 levels

Hi, I tried several settings by using the "family=gaussian" in "gl1ce", but none of them works. For the case "glm" can work. Here is the error message I got: > glm(Petal.Width~Sepal.Length+Sepal.Width+Petal.Length ,data=iris,family=gaussian()) > gl1ce(Petal.Width~Sepal.Length+Sepal.Width+Petal.Length ,data=iris,family=gaussian()) Error in eval(expr, envir,

Dear all, I have the following dataset: each row corresponds to count of forest floor small mammal captured in a plot and vegetation characteristics measured at that plot > sotr plot cnt herbc herbht 1 1A1 0 37.08 53.54 2 1A3 1 36.27 26.67 3 1A5 0 32.50 30.62 4 1A7 0 56.54 45.63 5 1B2 0 41.66 38.13 6 1B4 0 32.08 37.79 7 1B6 0 33.71 30.62

X is a matrix and F is a vector. F2 <- data.frame(cbind(X,F)) F2 V1 V2 V3 F 1 -0.250536332 -1.4755883 1.9580974 -2.136487 2 -0.009856084 0.4953269 0.5486092 -2.744482 3 -0.406962682 0.7729631 0.1861905 -2.891821 4 1.938780097 0.7469251 1.2537781 -1.212992 5 -0.332370358 1.1943637 0.7114278 -1.830441 modF<-formula(F ~ V1 + V2 + V3) #no error message

Hi there fellow R-users, Can anyone tell me how to build a prediction interval for a gaussian log-link model for the reponse variable?? I can find the standard error of the predictions but I cant seem to find the prediction interval. Is there a way I can calculate the prediction interval from the standard errors?? Here's the example: logX<-rnorm(100)

glm(S ~ -1 + Mdif, family=quasipoisson(link=identity), start=strt, sdat) gives error: > Error in glm.fit(x = X, y = Y, weights = weights, start = start, etastart > = > etastart, : > cannot find valid starting values: please specify some strt is set to be the coefficient for a similar fit glm(S ~ -1 + I(Mdif + 1),... i.e. (Mdif + 1) is a vector similar to Mdif. The error

Hello: Are there any libraries that will do a subset selection for glm's? I looked through leaps, but seems like it is specifically for linear regressions. Thank you. -Dhiren