similar to: Re: glm and data.frames (PR#108)

Full_Name: Glenn Stone Version: 1.5.1 and 1.6.0 OS: win2000 Submission from: (NULL) (168.140.227.9) The following code produces incorrect fitted values in version 1.5.1 and an error in 1.6.0 Error in "contrasts<-"(*tmp*, value = "contr.treatment") : contrasts apply only to factors In addition: Warning message: variable ihalf is not a factor in:

Hello - I cannot get to www.alpha.luc.ac.be/~jlindsey/rcode.html<http://www.alpha.luc.ac.be/%7Ejlindsey/rcode.html>to obtain and install the repeated library for use of glmm(). Is the web page not active? Can you give me an alternative location to obtain the repeated library? Thank you, Becky Parker [[alternative HTML version deleted]]

Hello, All: ????? Consider: Browse[2]> set.seed(1) Browse[2]> rpois(9, 1e10) NAs produced[1] NA NA NA NA NA NA NA NA NA ????? Should this happen? ????? I think that for, say, lambda>1e6, rpois should return rnorm(., lambda, sqrt(lambda)). ????? For my particular Monte Carlo, I have replaced my call to rpois with a call to the following: ?rpois. <- function(n,

it is easy to fit a distribution using fitdistr poisdata <- rpois(n = 100, lambda = 2) poismle <- fitdistr(poisdata, "Poisson") poismle but i would like to know whether its possible to get an identical result using glm. I use poistab <- data.frame(table(poisdata)) colnames(poistab) <- c("width","freq"); poistab[,"width"] <-

Crazy thought, but being that a sum of Poissons is Poisson in the sum, can you break your ?big? simulation into the sum of a few smaller ones? Or is the order of magnitude difference just too great? On Sun, Jan 19, 2020 at 1:58 PM Spencer Graves <spencer.graves at prodsyse.com> wrote: > This issue arose for me in simulations to estimate confidence, > prediction, and tolerance

On 1/20/20 12:33 PM, Martin Maechler wrote: > > It's really something that should be discussed (possibly not > here, .. but then I've started it here ...). > > The NEWS for R 3.0.0 contain (in NEW FEATURES) : > > * Functions rbinom(), rgeom(), rhyper(), rpois(), rnbinom(), > rsignrank() and rwilcox() now return integer (not double) > vectors.

Floor (maybe round) of non-negative numerics, though. Poisson should never have anything after decimal. Still think it?s worth allowing long long for R64 bit, just for purity sake. Avi On Sun, Jan 19, 2020 at 4:38 PM Spencer Graves <spencer.graves at prodsyse.com> wrote: > > > On 2020-01-19 13:01, Avraham Adler wrote: > > Crazy thought, but being that a sum of Poissons is

Technically, lambda can always be numeric. It is the observations which must be integral. Would hitting everything larger than maxint or maxlonglong with floor or round fundamentally change the distribution? Well, yes, but enough that it would matter over process risk? Avi On Sun, Jan 19, 2020 at 11:20 AM Benjamin Tyner <btyner at gmail.com> wrote: > So imagine rpois is changed, such

Dear All, The std. error of the estimated coefficients obtained by the summary.lm function can be calculated as: y=rnorm(20) x=y+rnorm(20) fit <- lm(y ~ x) summary(fit) sqrt( sum(fit$resid**2)/fit$df.resid * solve(t(model.matrix(fit))%*%model.matrix(fit)) ) Is posible calculate Std. Error for glm as lm, using cov(hat beta) = phi * solve(t(X) %*% hat W %*% X)^-1 on R? Who is hat W and

What kind of special magic does glm have? I'm working on a logistic regression solver (L-BFGS) in c and I've been using glm to check my results. I came across a data set that has a very high condition number (the data matrix transpose the data matrix) that when running my solver does not converge, but the same data set with glm was converging ( I love R :) ). I noticed that glm using

>>>>> Spencer Graves >>>>> on Sun, 19 Jan 2020 21:35:04 -0600 writes: > Thanks to Luke and Avi for their comments.? I wrapped "round" around the > call to "rnorm" inside my "rpois.".? For "lambda" really big, that > "round" won't do anything.? However, it appears to give integers in

R uses the C 'int' type for its integer data and that is pretty much universally 32 bit these days. In fact R wont' compile if it is not. That means the range for integer data is the integers in [-2^31, +2^31). It would be good to allow for a larger integer range for R integer objects, and several of us are thinking about how me might get there. But it isn't easy to get right, so

Craig, Thanks for your follow-up note on using the asypow package. My problem was not only constructing the "constraints" vector but, for my particular situation (Poisson regression, two groups, sample sizes of (1081,3180), I get very different results using asypow package compared to my other (home grown) approaches. library(asypow) pois.mean<-c(0.0065,0.0003) info.pois <-

On 2020-01-19 09:34, Benjamin Tyner wrote: >> ------------------------------------------------------------------------ >> Hello, All: >> >> >> ? ????? Consider: >> >> >> Browse[2]> set.seed(1) >> Browse[2]> rpois(9, 1e10) >> NAs produced[1] NA NA NA NA NA NA NA NA NA >> >> >> ? ????? Should this happen? >>

So imagine rpois is changed, such that the storage mode of its return value is sometimes integer and sometimes numeric. Then imagine the case where lambda is itself a realization of a random variable. Do we really want the storage mode to inherit that randomness? On 1/19/20 10:47 AM, Avraham Adler wrote: > Maybe there should be code for 64 bit R to use long long or the like? > > On

Can anyone explain to me why the AIC values are so different when using glm.nb and glm with a negative.binomial family, from the MASS library? I'm using R 1.8.1 with Mac 0S 10.3.4. >library(MASS) > dfr <- data.frame(c=rnbinom(100,size=2,mu=rep(c(10,20,100,1000),rep(25,4))), + f=factor(rep(seq(1,4),rep(25,4)))) > AIC(nb1 <- glm.nb(c~f, data=dfr)) [1] 1047 >

????? This issue arose for me in simulations to estimate confidence, prediction, and tolerance intervals from glm(., family=poisson) fits embedded in a BMA::bic.glm fit using a simulate.bic.glm function I added to the development version of Ecfun, available at "https://github.com/sbgraves237/Ecfun".? This is part of a vignette I'm developing, available at

Maybe there should be code for 64 bit R to use long long or the like? On Sun, Jan 19, 2020 at 10:45 AM Spencer Graves <spencer.graves at prodsyse.com> wrote: > > > On 2020-01-19 09:34, Benjamin Tyner wrote: > >> ------------------------------------------------------------------------ > >> Hello, All: > >> > >> > >> Consider:

Full_Name: Carmen Fernandez Version: 1.3.1 OS: Submission from: (NULL) (138.251.202.115) I think there is a problem when computing Pearson residuals, in that they seem to be computed at the raw residuals divided by the square root of the corresponding diagonal element of the weight matrix W evaluated at the last step of the iterative model fitting procedure (IWLS), instead of dividing by the

Thanks to Luke and Avi for their comments.? I wrapped "round" around the call to "rnorm" inside my "rpois.".? For "lambda" really big, that "round" won't do anything.? However, it appears to give integers in floating point representation that are larger than .Machine$integer.max.? That sounds very much like what someone would want.?