similar to: problems with optim, "for"-loops and machine precision

Hi all, I am trying to find estimates for 7 parameters of a model which should fit real data. I have a function for the negative log likelihood (NLL) of the data. With optim(method="L-BFGS-B",lower=0) I am now minimizing the NLL to find the best fitting parameters. My problem is that the algorithm does not converge for certain data sets. I have read that one should scale the fn

Dear all, I am trying to evaluate the optimisation behaviour of a function. Originally I have optimised a model with real data and got a set of parameters. Now I am creating simulated data sets based on these estimates. With these simulations I am estimating the parameters again to see how variable the estimation is. To this end I have written a loop which should generate a new simulated data

Hello, I am trying to use nlm to estimate the parameters that minimize the following function: Predict<-function(M,c,z){ + v = c*M^z + return(v) + } M is a variable and c and z are parameters to be estimated. I then write the negative loglikelihood function assuming normal errors: nll<-function(M,V,c,z,s){ n<-length(Mean) logl<- -.5*n*log(2*pi) -.5*n*log(s) -

Hi guys, I need some help to analyzing my data. I start to describe my data: I have 21 matrices, every matrix on the rows has users and on columns has items, in my case films. Element of index (i, j) represent the rating expressed by user i about item j. I have a matrix for each of professions. An example of a this type of matrix is: item 1 item 2 item 3 item4 id

Hi everyone! I am using the mle {stats4} to estimate the parameters of distributions by MLE method. I have a problem with the examples they provided with the mle{stats4} html files. Please check the example and my question below! *Here is the mle html help file * http://stat.ethz.ch/R-manual/R-devel/library/stats4/html/mle.html http://stat.ethz.ch/R-manual/R-devel/library/stats4/html/mle.html

for learning purposes and also to help someone, i used roger peng's document to get the mle's of the gamma where the gamma is defined as f(y_i) = (1/gammafunction(shape)) * (scale^shape) * (y_i^(shape-1)) * exp(-scale*y_i) ( i'm defining the scale as lambda rather than 1/lambda. various books define it differently ). i found the likelihood to be n*shape*log(scale) +

Hello, I have a log likelihood function that I was able to optimize using mle2. I have two years of the data used to fit the function and I would like to fit both years simultaneously to test if the model parameter estimates differ between years, using likelihood ratio tests and AIC. Can anyone give advice on how to do this? My likelihood functions are long so I'll use the tadpole

You could declare a matrix much larger than you intend to use. This works with a few megabytes of data. It is not very efficient, so scaling up may become a problem. m22 <- matrix(NA, 1:600000, ncol=6) It does not work to add a new column to the matrix, as in you get an error if you try m22[ , 7] but convert to data frame and add a column m23 <- data.frame(m22) m23$x7 <- 12 The only

Dear R developers, I noticed a bug in the stats4 package, specifically in the confint method applied to ?mle? objects. In particular, when some ?fixed? parameters define the log likelihood, these parameters are stored within the mle object but they are not used by the ?confint" method, which retrieves their value from the global environment (whenever they still exist). Sample code: >

The matrix equivalent of x <- ... v <- ... x[length(x)+1] <- v is m <- ... r <- ... m <- rbind(m, r) or m <- ... k <- ... m <- cbind(m, c) A vector or matrix so constructed never has "holes" in it. It's better to think of CONSTRUCTING vectors and matrices rather than INITIALISING them, because always being fully defined is important. It

Hello R Help, I am trying solve an MLE convergence problem: I would like to estimate four parameters, p1, p2, mu1, mu2, which relate to the probabilities, P1, P2, P3, of a multinomial (trinomial) distribution. I am using the mle2() function and feeding it a time series dataset composed of four columns: time point, number of successes in category 1, number of successes in category 2, and

"It would be really really helpful to have a clearer idea of what you are trying to do." Amen! But in R, "constructing" objects by extending them piece by piece is generally very inefficient (e.g. https://r-craft.org/growing-objects-and-loop-memory-pre-allocation/), although sometimes?/often? unavoidable (hence the relevance of your comment above). R generally prefers to take

Hi, I'm using Samba version 3.0.14a-r2 on Gentoo Linux as a PDC for a classroom environment. The server and the classroom are in different subnets, but on the same local LAN. The broadcast messages are not forwarded, so I have enabled a single machine in the classroom subnet to act as a WINS proxy. I am using Ghost to image the classroom machines which uses sysprep to prepare the

On Thu, 11 Dec 2003, Jeffrey Chang wrote: > Hello everybody, > > I'm running R 1.8.1 on both Linux and OS X compiled with gcc 3.2.2 and > 3.3, respectively. The following call seems to freeze the interpreter > on both systems: > > chisq.test(matrix(c(233, 580104, 3776, 5786104), 2, 2), > simulate.p.value=TRUE) > > By freeze, I mean, the function call never

Hello everybody, I'm running R 1.8.1 on both Linux and OS X compiled with gcc 3.2.2 and 3.3, respectively. The following call seems to freeze the interpreter on both systems: > chisq.test(matrix(c(233, 580104, 3776, 5786104), 2, 2), simulate.p.value=TRUE) By freeze, I mean, the function call never returns (running > 10 hours so far), the process is unresponsive to SIGINT (but I

This is obviously not a long term acceptable solution. Could someone please enlighten me as to the permanent solution, or at least see why it doesn't work? This is not my problem, therefore I have not had the opportuntity to test - I don't have a version that old. Let me know if it is solved in a newer version as well. Thanks. It is a workable solution, although the generally accepted

I am using the Ben bolker's R package "bbmle" to estimate the parameters of a binomial mixture distribution via Maximum Likelihood Method. For some data sets, I got the following warning messages: *Warning: optimization did not converge (code 1: ) There were 50 or more warnings (use warnings() to see the first 50)* Also, warnings() results the following: *In 0:(n - x) : numerical

Hi Denis, I just tried this but still I can't log on the samba server with a domain user! If I try to do so I get the error: [2003/10/29 08:48:37, 0] auth/auth_util.c:make_server_info_info3(1017) make_server_info_info3: pdb_init_sam failed! in the log file of the client on samba server... Is there anytihng else I have to adjust on the samba server? I sucessfully joined the domain with

Dear R help, I am trying solve an MLE convergence problem: I would like to estimate four parameters, p1, p2, mu1, mu2, which relate to the probabilities, P1, P2, P3, of a multinomial (trinomial) distribution. I am using the mle2() function and feeding it a time series dataset composed of four columns: time point, number of successes in category 1, number of successes in category 2, and

Hi, The following distribution is known as Kumaraswamy binomial Distribution. http://r.789695.n4.nabble.com/file/n4636782/kb.png For a given data I need to estimate the paramters (alpha and beta) of this distribution(Known as Kumaraswamy binomial Distribution, A Binomial Like Distribution). For that, in order to use *optim()*, I first declared the Negative Log-likelihood of this distribution as