similar to: polr() error message wrt optim() and vmmin

Hi, I'm running gam (mgcv version 1.3-29) and glm (logit) (stats R 2.61) on the same models/data, and I got error messages for the gam() model and warnings for the glm() model. R-help suggested that the glm() warning messages are due to the model perfectly predicting binary output. Perhaps the model overfits the data? I inspected my data and it was not immediately obvious to me (though I

Hi all. I'm just starting to explore ordinal multinomial regression. My dataset is 300,000 rows, with an outcome (ordinal factor from 1 to 9) and five independent variables (all continuous). My first stab at it was this: pomod <- polr(Npf ~ o_stddev + o_skewness + o_kurtosis + o_acl_1e + dispersal, rlc, Hess=TRUE) And that worked; I got a good model fit. However, a variety of other

In polr.R the (several) functions gmin and fmin contain the code > theta <- beta[pc + 1L:q] > gamm <- c(-100, cumsum(c(theta[1L], exp(theta[-1L]))), 100) That's bad. There's no reason to suppose beta[pc+1L] is larger than -100 or that the cumulative sum is smaller than 100. For practical datasets those assumptions are frequently violated, causing the

I am calling 'vmmin' several times from a C function (which is called via .C). It works very well, except for memory consumption. The cause is that vmmin allocates memory via R_alloc, and this memory is not freed as vmmin exits. Instead all the allocated memory is freed on return of the .C call. In one application, I have 2000 functions of 500 variables each to minimize. In each call to

Error in optim(method = "BFGS", c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, : initial value in 'vmmin' is not finite I am running a logit model with latent class segments. I successfully got estimates for 2 segments. However when I tried to increase the no. of segments, I got this error message at the end. I checked my code again but can't find anything wrong. Is this error

Dear r helpers, I'm trying to run the models of the BIOMOD package and got this message: * Error in nnet.default (x, y, w,...): **initial value in 'vmmin' is not finite* What does this mean? and how can I correct it? Thank you *Andreia* [[alternative HTML version deleted]]

I am writing code in C and would like to call R's functions samin and vmmin (optimization routines: simulated annealing and BFGS) I do not understand how to create and pass in the function (as well as the extra arguments it needs) I am optimizing. I have read the R Extensions manual but it is still unclear to me. Could you give me some pointers and/or direct me to some example code which

Dear r-helpers, I'm experiencing some problems in fitting a maximum likelihood binomial model to some of my data. The error is in optim, which founds: Error in optim(par = c(0.2, 0.5), fn = function (p) : initial value in 'vmmin' is not finite Yes, I know it's a common problem, and I've carefully searched and readed all the posts about the issue, But, I can't find a

Dear listers, Am currently using MCMC approaches to estimate some parameters of my model. One parameter has to be updated using a tuned gamma distribution. So at each iteration I estimate the mean and variance of the density of the gamma approximation using "vmmin" (i also supply the gradient argument). For moderate replications the procedure works, but if I increase them R crashes.

What is the current state of the R-art for ordered probit models, and more esoterically is there any available R strategy for ordered cauchit models, i.e. ordered multinomial alternatives with a cauchy link function. MCMC is an option, obviously, but for a univariate latent variable model this seems to be overkill... standard mle methods should be preferable. (??) Googling reveals that spss

[This email is either empty or too large to be displayed at this time]

Dear Rhelpers I have the problem with initial values, could you please tell me how to solve it? Thank you June > p = summary(maxLik(fr,start=c(0,0,0,1,0,-25,-0.2))) Error in maxRoutine(fn = logLik, grad = grad, hess = hess, start = start, : NA in the initial gradient > p = summary(maxLik(fr,start=c(0,0,0,1,0,-25,-0.2),method="BFGS")) Error in optim(start, func, gr =

Dear r helpers I run the following code for nested logit and got a message that Error in optim(c(0, 0, 0, 0, 0.1, -2, -0.2), fr, hessian = TRUE, method = "BFGS") : initial value in 'vmmin' is not finite What does this mean? and how can I correct it? Thank you June > yogurt = read.table("yogurtnp.csv", header=F,sep=",")> attach(yogurt)>

I am trying to use the lnam autocorrelation model from the SNA package. I have it running for smaller adjacency matrices (<1,500) it works just fine but when my matrices are bigger 4000+. I get the error: > lnam1_01.adj<- lnam(data01$adopt,x01,ec2001.csr) Error in optim(rho, n2ll.rho, method = method, control = control, beta = parm$beta, : initial value in 'vmmin' is not

Full_Name: Henric Nilsson Version: 2.3.0 Patched (2006-05-09 r38014) OS: Windows 2000 SP4 Submission from: (NULL) (83.253.9.137) When supplying an unavailable link to `quasibinomial', the error message looks strange. E.g. > quasibinomial("x") Error in quasibinomial("x") : 'x' link not available for quasibinomial family, available links are "logit",

Dear all, A am trying to fit a generalized linear mixed effects model with a binomial link function, my response data is binary, using the lme4 R package, for the glmer model but with the cauchit link function (CDF of Cauchy distribution), under the package this has not yet been coded and was wondering if anyone knew a way in which I could incorporate this link function into the code. Thankyou

This is not so much a bug as an infelicity in the code that can easily be fixed. The initialize expression in the quasi family function is, (uniformly for all links and all variance functions): initialize <- expression({ n <- rep.int(1, nobs) mustart <- y + 0.1 * (y == 0) }) This is inappropriate (and often fails) for variance function "mu(1-mu)".

Hi all!! I am using a proportinal odds model to study some ordered categorical data. I am trying to predict one ordered categorical variable taking into account only another categorical variable. I am using polr from the R MASS library. It seems to work ok, but I'm still getting familiar and I don't know how to assess goodness of fit. I have this output, when using response ~ independent

Hi all!! I am using a proportinal odds model to study some ordered categorical data. I am trying to predict one ordered categorical variable taking into account only another categorical variable. I am using polr from the R MASS library. It seems to work ok, but I'm still getting familiar and I don't know how to assess goodness of fit. I have this output, when using response ~ independent

Dear Sir, I get an error message when I use polr() in MASS package. My data is "ord.dat". I made "y" a factor. y y1 y2 x lx 1 0 0 0 3.2e-02 -1.49485 2 0 0 0 3.2e-02 -1.49485 3 0 0 0 1.0e-01 -1.00000 4 0 0 0 1.0e-01 -1.00000 5 0 0 0 3.2e-01 -0.49485 6 0 0 0 3.2e-01 -0.49485 7 1 1 0 1.0e+00 0.00000 8 0 0 0 1.0e+00 0.00000 9 1 1 0