similar to: How to define stopping criterium for Optim with L-BFGS-B

Hello, I've just a little problem using the function optim. Here is the function I want to optimize : test_function(x){(exp(-0.06751 + 0.25473*((x[1]-350)/150) + 0.04455*((x[2]-40)/20) + 0.09399*((x[3]-400)/100) - 0.17238*((x[4]-250)/50)- 0.45984*((x[5]-550)/150)-0.39508*((x[1]-350)/150)* ((x[1]-350)/150) - 0.05116*((x[2]-40)/20)* ((x[2]-40)/20) - 0.27735*((x[3]-400)/100)*((x[3]-400)/100) -

Hello Everybody, I have a problem with the output of lm. In fact, when I compare the R output and the Splus output, the results appears to be different : I have a dataframe with 5 factors , 16 rows and a response. First, I tried to fit this response to a linear model : lm(rep~(A+B+C+D+E),data). The results beetween SPlus and R are different. In a second time, I tried a new fit :

Hello, I am try to perform a modeling which is relevant in a strongly heteroscedastic context. So I perform a dual modeling (modeling of both mean and variance of a response) in using the following loop: jointmod <- function(formula, data, itercrit=10,devcrit=0.0001) { # # Init step # init <- glm(formula=formula,family=gaussian, data=data) response <-

Hello, I try to install the R111 version for few days on my PC (on Windows NT), and every time the same problem : I start the rwin.exe application and a couple of seconds later, the install failed, and my PC is down. I must then swich it down and restart a new session. If someone has some help ... Thanks -- Isabelle Zabalza-Mezghani Tel : 01 47 52 61 99 Institut Fran?ais du P?trole

Hello, I would like to know if there is a possibility to "clean" the R memory during a R session. In fact, I realize a lot of instruction with large objects (matrix of 500*5000), and I can not manage to achieve the end of my script due to memory lack. Of course, I've tried to remove all "temporary object" during the script execution and to perform a garbage collector ...

Hello everybody, I'm writting some functions for experimental designs. The one I'm working on is similar to "fac.design" on Splus. The problem I have is with the form of the output : When the argument fraction is submitted, Splus gives something like : A B C 1 - - - 2 + + - 3 + - + 4 - + + Fraction : ~A:B:C The first part of this output is the

Hello, I would have some details and explanations about the results I get. In fact, I start with a uniform sample between -1 and 1, and then plot its density. My problem is that the density ranges are much more longer than I expected : samp <- runif(10000,-1,1) plot(density(samp)) Instead of varying between -1 and 1, the density varies between approximaly -1.5 and 1.5 Could someone explain

Hello R-users, I would like to install the Java package delivered by the Omegahat project. I'm working on a Windows NT4.0 workstation. Can someone explain me ( in details please) the way to install this package. I've already read the doc relative to this topic, but I'm afraid to be too bad in Informatic to understand really well the install process. I hope someone has an answer for

Hello, I wonder if anybody has some idea about how to solve my problem : I am working , I would say trough an experimental design approach (perform experiments, get responses, make regression, sensitivity analysis, risk analysis ...). The problem is now that I have to face with not only a response but a vector or a matrix (typically a spatial distribution of a physical property ... pressure).

Hello, I would like to know if there is a possibility to open an R session via Java (using the SJava package), then to terminate it, and re-run another. It seems not to be possible. If this is the case, I would like to understand where is the problem or the limitation (is it due to the SJava implementation, to the Java behavior, or to the R application). In fact, I am interesting in re-starting

Hi, I have some doubt about how qualitative factors are coded in R. For instance, I consider a response y, a quantitative factor x and a qualitative factor m at 3 levels, generated as follow : y_c(6,4,2.3,5,3.5,4,1.,8.5,4.3,5.6,2.3,4.1,2.5,8.4,7.4) x_c(3,1,3,1,2,1,4,5,1,3,4,2,5,4,3) m_gl(3,5) lm(y~x+m) Coefficients: (Intercept) x m2 m3 3.96364 0.09818

Hi, I would like to have some advise about how to install Dopt on R for Windows. I can not manage to properly compile Dopt.f anf to get a suitable dll for R. Can someone explain me step by step what should I do ? Thanks in advance Isabelle Zabalza-Mezghani

Hello, I've just fitted a model with multi-responses, and I get an object of class "lm" "mlm". My problem is that as soon as I invoke the predict method for a dataframe "newdata", the methods runs and give me back prediction at the fitting points but not for newdata. Does someone has an explanation for this behavior, and some ideas to make predict.mlm work

Hi, I have a problem with the memory size within R. I would like to know if there is any may for getting back free memory during a R session. I've tried with rm and gc but it still craches. I am working on windows R1.3.1. I know I have big datasets, but during my study, some results are temporary, and I would like to get back the memory allocated to this temporary object as soon as I remove

Hi everybody, I'm trying to use fortran subroutines in a R function. I've generated a DLL file using my fortran subroutines, and I've got a DLL which includes an export table with the names of my subroutines. Then, I use the R command library(test), where test is a directory located in the R library tree and which contains a R directory (with my R functions) and a libs directory (with

Hello, I would like to know where I can find a Windows binary version of R 1.6.2, since I want to test the last SJava library (which has been built under 1.6.2) which does not work on R 1.7.1 I've already search on the CRAN web site but on the r1070 page, if I click on 'old', I get an error. Thanks in advance, Isabelle Isabelle Zabalza-Mezghani IFP - Reservoir Engineering

Hello, I would like to treat a very simple case : to fit a linear model with 5 parameters (including main terms, interactions and quadratic terms using a central composite design with 27 runs) for a constant response (e.g resp = 100.0). The fitting process works and return me a good intercept value (the value of my constant) and some negligeable effects (around e-15). But, I don't understand

This example is caused by R's messing with formula environments. That's explained in ?formula, but should it not be explained in ?model.frame ? Simple test: data <- data.frame(y=rnorm(100), x=1:100) testit <- function(formula) { weights <- runif(100) glm(formula, weights=weights, data=data) } testit(y ~ x) weights is looked for in the environment of the formula, not of

Hello, I want to call glm inside a function. In the first lines of code I build the weights, the formula, ... and then I call glm with the following command: glm(formularesp, data=data, family=familyresp, weights=eval(data$weights) My problem is that the fitting proccess is performed just like if weights=NULL, even if my weights are not equals to 1. I've performed some return() command

Hello, I am using the function "lm" to fit several responses at the same time (100 responses). At the end of the fit, I get an object of class "mlm". I would like to know if there is a way to access to each of the 100 underlying models separately (is it a list, ... ?). Which syntax should I use to see and use the 15th model (for instance) just like it is possible for classical