similar to: How does logLik(lm(...)) find the maximum log likelihoods

In my own package, I want to use the default S3 method of the generic function lrtest() from the lmtest package. Since I need only one function from lmtest, I tried to use importFrom in my NAMESPACE: importFrom(lmtest, lrtest) However, this fails R CMD check in the examples: Error in UseMethod("lrtest") : no applicable method for 'lrtest' applied to an object of class

Hi, I'm trying to extract a histogram over the results from a bootstrap. However I keep receiving the error message "Error in hist.default(boot.lrtest$ll, breaks = "scott") : 'x' must be numeric". The bootstrap I'm running looks like: > boot.test <- function(data, indeces, maxit=20) { + y1 <- fit1+e1[indeces] + mod1 <- glm(y1 ~ X1-1, maxit=maxit) +

Hi. I would like to use a likelihood-ratio test to compare a linear model (lm()) to another linear model containing the first one to see if the extra factors are needed - but I was unable to find any help on how to do that. Could you help me and tell me what to do? Or tell me where to find help on this topic? Many thanks in advance! Lars

Dear R-experts: I am using survreg() to estimate the parameters of a Weibull density having right-censored observations. Some observations are weighted. To do that I regress the weighed observations against a column of ones. When I enter the data as 37 weighted observations, the parameter estimates are exactly the same as when I enter the data as the corresponding 70 unweighted observations.

Hello all. I was wondering if the logLike.nls() and logLike.nlme() functions are still being used. Neither function seems to be available in the most recent release of R (1.9.1). The following is contained in the help file for logLik(): "classes which already have methods for this function include: 'glm', 'lm', 'nls' and 'gls', 'lme' and others in

Does the ?logLik? function applied to a ?glm? and ?glm.nb? (from MASS package) calculate the complete log-likelihoods, or does it drop the constant terms of the equation? (It?s not clear from the associated help pages, and I?ve found no reference from searching the R help mailing list) Thank you, Kelly Young

Hi, I'm trying to compare models, one of which has all parameters fixed using offsets. The log-likelihoods seem reasonble in all cases except the model in which there are no free parameters (model3 in the toy example below). Any help would be appreciated. Cheers, Jarrod x<-rnorm(100) y<-rnorm(100, 1+x) model1<-lm(y~x) logLik(model1) sum(dnorm(y, predict(model1),

Hi group, I found a module for adaptive kernel density estimation for Stata users, but unfortunetly I don't have access to Stata, can I find a similar approach using R? Thank u so much 4 ur time. [[alternative HTML version deleted]]

It appears that the degrees of freedom reported by logLik changed between R 1.4.1 and R 1.5.1. Is this true? Detail: > I have been using the lm and logLik functions in R to develop code using > version 1.4.1. When I run it on version 1.5.1, I'm getting different > degrees of freedom with the logLik function. Version 1.5.1 seems to give > one extra degree of freedom than

After some struggling with the data format, non-standard in BioConductor, I have gotten cor.balance in package CORREP to work. My desire was to obtain maximum-likelihood p-values from the same data object using cor.LRtest, but it appears that this function wants something different, which I can't figure out from the documentation. Briefly, my dataset consists of 36 samples from 12

Hello all, I was wondering if someone can enlighten me as to the difference between the logLik in R vis-a-vis Stata for a GLM model with the gamma family. Stata calculates the loglikelihood of the model as (in R notation) some equivalent function of -1/scale * sum(Y/mu+log(mu)+(scale-1)*log(Y)+log(scale)+scale*lgamma(1/scale)) where scale (or dispersion) = 1, Y = the response variable, and mu

Dear UseRs, I wrote the following function to use MLE. --------------------------------------------- mlog <- function(theta, nx = 1, nz = 1, dt){ beta <- matrix(theta[1:(nx+1)], ncol = 1) delta <- matrix(theta[(nx+2):(nx+nz+1)], ncol = 1) sigma2 <- theta[nx+nz+2] gamma <- theta[nx+nz+3] y <- as.matrix(dt[, 1], ncol = 1) x <- as.matrix(data.frame(1,

I am cleaning up the rms package to not export functions not to be called directly by users. rms uses generic functions defined in other packages. For example there is a latex method in the Hmisc package, and rms has a latex method for objects of class "anova.rms" so there are anova.rms and latex.anova.rms functions in rms. I use:

Hello, I'm trying to use AIC to choose between 2 models with positive, continuous response variables and different error distributions (specifically a Gamma GLM with log link and a normal linear model for log(y)). I understand that in some cases it may not be possible (or necessary) to discriminate between these two distributions. However, for the normal linear model I noticed a discrepancy

Dear all, I have a very basic question: how does the logLik function work for poisson models? Example: I simulate 20 observations from a Poisson distribution with mean 800. y <- rpois(20,800) model <- glm(y ~ 1, family=poisson()) logLik(model) I would like to know what's the exact formula the function logLik uses. I looked at ?extractAIC but I cannot sort it out. Can you please

Hello everyone, I hope this is a fair enough question, but I don’t have access to a copy of Bates and Pinheiro. It is probably quite obvious but the answer might be of general interest. If I fit a fixed effect with an added quadratic term and then do it as an orthogonal polynomial using maximum likelihood I get the expected result- they have the same logLik.

Dear all, I'd like to discuss about a possible bug in function StructTS of stats package. It seems that the function returns wrong value of the log-likelihood, as the added constant to the relevant part of the log-likelihood is misspecified. Here is an simple example: > data(Nile) > fit <- StructTS(Nile, type = "level") > fit$loglik [1] -367.5194 When computing the

List: I'm helping a colleague with some Poisson regression modeling. He uses SAS proc GENMOD and I'm using glm() in R. Note on the SAS and R output below that our estimates, standard errors, and deviances are identical but what we get for likelihoods differs considerably. I'm assuming that these must differ just by some constant but it would be nice to have some confirmation

I have a question about the computation of the degrees of freedom in a linear model: x <- runif(20); y <- runif(20) f <- lm(y ~ x) logLik(f) 'log Lik.' -1.968056 (df=3) The 3 is coming from f$rank + 1. Shouldn't it be f$rank? This affects AIC(f). Thanks Frank ----- Frank Harrell Department of Biostatistics, Vanderbilt University -- View this message in context:

I have a dataset that I am running non-linear regression on via the following code: Hill <- function(E0,Em,C50,g,C){ # # Hill is the hill interaction function. # # E0 Represents the minimum interaction Effect # # Em Represents the Maximum Interaction Effect # # C50 represents the concentration at which 50% of the effect occurs. # # gamma represents the cooperativity of the