similar to: residuals in multinom

Hi, I am searching for methods to compare regression models with an ordered categorical response variable (polr versus multinom). The pattern of predictions of both methods (using the same predictor variables) is quite different and the AIC is smaller for the multinom approach. I guess polr has more strict premises for the structure of the response variable, which methods can be used to test for

Dear all, I've got a list L <- list(L1 = list (foo = "bar" , SL = NULL ) , L2 = list ( foo = "bar" , SL = list (SSL1 = list (DF = data.frame(val = 21, foo = "bar") , DFOO = list(foo = "foo", bar = "bar") ) ,

Hi I am using nlsList from the nlme package to estimate group specific coefficient sets to get a first impression of the structure of coefficients in relation to potential predictor variables. In some cases I got coefficients in other there are conversion problems due to bad start values I guess, resulting in typical error messages as known from nls: Error in nls(formula = formula, data = data,

Hello, sorry for writing here because my problem is not a realt bug but may be a solution for many people working with R: I miss the feature for commenting some lines of code at once without writing a bunch of "#" in front of each line. This is interesting for trying out some code. I found some workarounds like IFELSE(FALSE){} and `!`<- function(x) { if (inherits(x,

Dear All: I am interested in multinomial logit models (function multinon, library nnet) but I'm having troubles in choose whether to define the predictors as factors or not. I had posted earlier this example (thanks for the reply ronggui): worms<- data.frame(year= rep(2000:2004, c(3,3,3,3,3)),age=rep(1:3,5),

Dear R users, Does anyone know if it is possible to use multinom to do a polychotomous fit using one categorical and one numeric variable as response. The doc. for multinom states that for formula , response can be K>2 classes. Is this 2 and more, or as I have understood it only greater than 2. I have tried fitting my data, but have only encountered error messages. On another note, Is it

Hi, I found that using different contrasts (e.g. contr.helmert vs. contr.treatment) will generate different fitted probabilities from multinomial logistic regression using multinom(); while the fitted probabilities from binary logistic regression seem to be the same. Why is that? and for multinomial logisitc regression, what contrast should be used? I guess it's helmert? here is an example

I want to get the logLik to calculate McFadden.R2 ,ML.R2 and Cragg.Uhler.R2, but the value from multinom does not have logLik.So my quetion is : is logLik meaningful to multinomial logistic model from multinom?If it does, how can I get it? Thank you! ps: I konw VGAM has function to get the multinomial logistic model with logLik, but I prefer use the function from "official" R

I have RedHat linux 9 with R 1.8. I'm estimating models with multinom with a dependent variable that has 3 different values. Sometimes the models run fine and I can understand the results. Sometimes when I put in another variable, I see an indication that the estimation did work, but then I can't get the summary method to work. It's like this: > votemn1 <-

Dear R users, I'm using a multinomial LOGIT model to analyse choice behaviour of consumers (as part of my masters thesis research). Using the R documentation and search on the R website I have a working script now. Parameters are estimated and I can access them via coefficients(multinom.out). In order to see if the parameters are significant I like to access the standard errors in the

Best R Help, I like to estimate a Multinomial Logit Model with 10 Classes. The problem is that the number of observations differs a lot over the 10 classes: Class | num. Observations A | 373 B | 631 C | 171 D | 700 E | 87 F | 249 G | 138 H | 133 I | 162 J | 407 Total: 3051 Where my data looks like: x1 x2 x3 x4 Class 1 1,02 2 1 A 2 7,2 1 5 B 3 4,2 1 4 H 1 4,1 1 8 F 2 2,4 3 7 D 1 1,2 0 4 J 2 0,9

Hi all, I'm working on a multinomial (or "polytomous") logistic regression using R and have made great progress using multinom() from the nnet library. My response variable has three categories, and there are two different possible predictors. I'd like to use the likelihoods of certain models (ie, saturated, fitteds, and null) to calculate Nagelkerke R-squared values for

Dear all, I have found some difference of the results between multinom() function in R and multinomial logistic regression in SPSS software. The input data, model and parameters are below: choles <- c(94, 158, 133, 164, 162, 182, 140, 157, 146, 182); sbp <- c(105, 121, 128, 149, 132, 103, 97, 128, 114, 129); case <- c(1, 3, 3, 2, 1, 2, 3, 1, 2, 2); result <- multinom(case ~ choles

Hi, I am having difficulty entering a 'programmable' argument into the multinom function from the nnet package. Interactively, I can get the function to work fine by calling it this way: z1=multinom(formula = class.ind(grp[-outgroup])~ (PC1 + PC2 + PC3), data=data.frame(scores)) However I need to be able to change the number of variables I am looking for in 'scores' and so am

Is anyone aware of a way to check whether multinom has converged by checking a component of the output ? I'm not familar with nnet but, since multinom calls nnet , maybe there is an extra argument once can send to multinom to capture this information. Thanks.

Dear all, I'm fitting a multinom function to my dataset (multinom(outcome~age+K+D)) and I need to present my results on a poster. Does someone know a nice way of doing that? I think I saw in an archive that you cannot plot a multinom.object, is it true? Thank you by advance for your help, Cheers Camille

Hi, I am trying to specify a multinomial logit model using the multinom function from the nnet package. Now I add another independent variable and it halves the AIC as given by summary(multinom()). But when I call Anova(multinom()) from the car package, it tells me that this added variable is insignificant (Pr(>Chisq)=0.39). Thus, the improved AIC suggests to keep the variable but the Anova

Good Day: When I used: multinom(formula = Y ~ X1 + X2 + X3 + X1:X2 + X1:X3 + X3:X2 + X1^2 + X2^2 + X3^2, data = DATASET), I get estimates and AIC for the model containing main effects and interactions only (no squared terms)...and FYI, all predictors are continuous. Is this "normal" behavior? If I run this in S-Plus I get estimates and AIC for the model containing all terms(including

Running a binary logit model on the data df <- data.frame(y=sample(letters[1:3], 100, repl=T), x=rnorm(100)) reveals some residual deviance: summary(glm(y ~ ., data=df, family=binomial("logit"))) However, running a multinomial model on that data (multinom, nnet) reveals a residual deviance: summary(multinom(y ~ ., data=df)) On page 203, the MASS book says that "here the

Hello folks, Let me first apologize: I'm not a professional nor a mathematician, just an ordinary guy, fooling around with the excellent R-package. I know the basic principles behind statistics, but haven't read anything more advanced than the ordinary first probability and statistics courses. Enough disclaimers? Good! I was examining the multinom-function (in the nnet-package) the other