similar to: R code for normal mixture EM algorithm

Please help me in writing the R code for this problem. I've been solving this for 4 days. It was hard for me to solve it. It's a simulation problem in R. The problem is My true model is a normal mixture which is given as 0.5 N(-0.8,1) + 0.5 N(0.8,1). This model has two components. I will get a random sample of size 100 from this model. I will do this 300 times. That means, I will have

Hi, I would like to know if it is possible to have a "R code" to estimate the parameters of a mixture of bivariate (or multivariate) normals via EM Algorithm. I tried to write it, but in the estimation of the matrix of variance and covariance, i have some problems. I generate two bidimensional vectors both from different distribution with their own vector means and variance and

I couldn't find a direct answer in CRAN to this question, so I'm asking with some trepidation. I have a multivariate dataset (data.frame) with columns that can be expressed as a set of mixed normals (at least I think) and need to impute values that have constraints (truncated mixture of normals where the values cannot be below zero). If there isn't a package that can do this, is there

Hi, I woul like to know if it is possible to have a "R code" to generate EM Algorithm for a normal bivariate mixture. Best regard, S.F.

Hi, I am currently working on fitting a mixture density to financial data. I have the following data: http://s000.tinyupload.com/?file_id=00083355432555420222 I want to fit a mixture density of two normal distributions. I have the formula: f(l)=πϕ(l;μ1,σ21)+(1−π)ϕ(l;μ2,σ22) my R code is: normalmix<-normalmixEM(dat,k=2,fast=TRUE) pi<-normalmix$lambda[1] mu1<-normalmix$mu[1]

Hi, i would to know, if someone have ever write the code to estimate the parameter (mixing proportion, mean, a var/cov matrix) of a mixture of two multivariate normal distribution. I wrote it and it works (it could find mean and mixing proportion, if I fix the var/cov matrix), while if I fix anything, it doesn't work. My suspect is that when the algorithm iterates the var/cov matrix, something

Hi, i would to know, if someone have ever write the code to estimate the parameter (mixing proportion, mean, a var/cov matrix) of a mixture of two multivariate normal distribution. I wrote it and it works (it could find mean and mixing proportion, if I fix the var/cov matrix), while if I fix anything, it doesn't work. My suspect is that when the algorithm iterates the var/cov matrix, something

Dear list, I have been comparing the outputs of two packages for latent class regression, namely 'flexmix', and 'mmlcr'. What I have noticed is that the flexmix package appears to come up with a much better fit than the mmlcr package (based on logLik, AIC, BIC, and visual inspection). Has anyone else observed such behaviour? Has anyone else been successful in using the mmlcr

Hi, All, I want to fit a normal mixture model. Which package in R is best for this? I was using the package 'mixdist', but I need to group the data into groups before fitting model, and different groupings seem to lead to different results. What other package can I use which is stable? And are there packages that can automatically determine the number of components? Thank you, Cindy

I've been trying with no success to model mixtures of Gamma distributions using the package flexmix (see examples below). Can anyone help me get it to model better? Thanks very much. -Ben ## ## Please help me get flexmix to correctly model mixtures of ## Gamma distributions. See examples below. ## library('flexmix') ## ## Plot a histogram of dat and the Gamma mixture model given

Hello all, I am trying to fit a truncated mixture model and I wrote a driver for flexmix following the example in the vignette, but it doesn't work for me: it assigns all data points to one component only, e.g.: > > source('bugged.R') > > Call: > flexmix(formula = x ~ 1, k = 2, model = truncatedmodel(lower = -4, > upper = 4)) > > prior size

Dear all, I am trying to model number of samples from a given series. The series are modelled according Gamma function. In order to estimate the # samples, I use BIC/AIC with MLE (computed from dgamma function). Here is the code I have. __BEGIN__ mlogl <- function( x_func, theta_func, samp) { # computing log_likelihood return( - sum(dgamma(samp, shape = x_func, scale=theta_func, log

Dear All, Suppose Z_i, i=1,...,m are marginally identically distributed as a two normal mixture p0*N(0,1) + (1-p0) *N( miu_i, 1) where miu_i are identically distributed according to a mixture and I have generated Z_i one by one . Now suppose these m random variables are jointly m-dimensional normal with correlation matrix M= (m_ij). How to proceed next or how to start correctly ? Question:

>>>>> "MM" == Martin Maechler <maechler at stat.math.ethz.ch> >>>>> on Tue, 18 May 2010 12:37:21 +0200 writes: >>>>> "GaGr" == Gabor Grothendieck <ggrothendieck at gmail.com> >>>>> on Mon, 17 May 2010 09:45:00 -0400 writes: GaGr> BIC seems like something that would logically go into stats

I'm using R 2.0.1. I looked in the email archives but didn't see anything on this topic. I've noticed a surprising (to me) difference between AIC and BIC: > methods("AIC") [1] AIC.default* AIC.logLik* > methods("BIC") [1] BIC.gls* BIC.lm* BIC.lme* BIC.lmList* BIC.logLik* BIC.nls* The BIC.gls BIC.lm BIC.lme BIC.lmList and BIC.nls functions appear

Dear List Members, I have encountered two problems when using the step function to select models. To better illustrate the problems, attached is an R image which includes the objects needed to run the code attached. lm.data.frame have factor variables with 3 levels. The following run shows the first problem. AICs (* and **) are different. I noticed that the Df for rs13482096:rs13483699 is 4,

Hello I am trying to find an automated way of fitting a mixture of normal and log-normal distributions to data which is clearly bimodal. Here's a simulated example: x.1<-rnorm(6000, 2.4, 0.6)x.2<-rlnorm(10000, 1.3,0.1)X<-c(x.1, x.2) hist(X,100,freq=FALSE, ylim=c(0,1.5))lines(density(x.1), lty=2, lwd=2)lines(density(x.2), lty=2, lwd=2)lines(density(X), lty=4) Currently i am using

Dear List Members, I have encountered two problems when using the step function to select models. To better illustrate the problems, an R image (step.add1.test.RData) which includes the objects needed to run the code (step.add1.test.R) can be found at www.biostat.wisc.edu/~pwang/r-help/<http://www.biostat.wisc.edu/%7Epwang/r-help/> lm.data.frame have factor variables with 3 levels. The

Hi, Does anyone know how to compute the quantile of a mixture of four bivariate normal distriutions? Many thanks! Hannah [[alternative HTML version deleted]]

Dear R helper, I hope that u can help me to sort out my problem because I sent an E-mail last night to R-list but I have not receive any help and at the same time I think this problem is not so hard. I have used the following functions before > K<-10 > prime<-c(2,3,5,7,11,13,17) > UN<-seq(1:K)%*%t(sqrt(prime)) > U1<-UN-as.integer(UN) > U<-matrix(qnorm(U1),K,7)