similar to: LME: internal workings of QR factorization --repost

Hi: I've been reading "Computational Methods for Multilevel Modeling" by Pinheiro and Bates, the idea of embedding the technique in my own c-level code. The basic idea is to rewrite the joint density in a form to mimic a single least squares problem conditional upon the variance parameters. The paper is fairly clear except that some important level of detail is missing. For

Hello, Currently, determinant(A) calculates the determinant of 'A' by factorizing A=LU and computing prod(diag(U)) [or the logarithm of the absolute value]. The factorization is done by LAPACK routine DGETRF, which gives a status code INFO, documented [1] as follows: *> INFO is INTEGER *> = 0: successful exit *> < 0: if INFO = -i, the i-th

Hello, I'm trying to fit a special kind of proportional odds model from: Whitehead et al. (2001). Meta-analysis of ordinal outcome using individual patient data. Statistics in medicine 20: 2243-2260. (model 2) The data are as follows: library(nlme) library(geepack) library(Design) library(MASS) options(contrasts=c("contr.SAS","contr.poly")) counts <-

Dear All, Apologies if this is too simple for this list. Let us assume that you have an instrument measuring particle distributions. The output is a set of counts {n_i} corresponding to a set of average sizes {d_i}. The set of {d_i} ranges from d_i_min to d_i_max either linearly of logarithmically. There is no access to further detailed information about the distribution of the measured sizes, but

Hi, I am drawing several plots and want to have italics in a main title; this is easy with expression(). However, I want also to add a value to it, say n_i, that depends on an ith plot. For this I am using paste(). An example: n_i = 10, 20, 30; I want to draw a plot for each i with the title: "Relative efficiency for sample size n = n_i", where n should be in italics, and of course n_i

Hello, I'm very new to using R, but I was told it could do what I want. I'm not sure how best to enter the information but here goes... I'm trying to transfer the following integral into R to solve for ln(gamma_1), on the left, for multiple instances of gamma_i and variable N_i. gamma_i is, for example, (0, 0.03012048, 0.05000000, 0.19200000, 0.44000000, 0.62566845) N_i (N_1 or

I am trying to create a set of wavelets in frequency space--namely Cauchy wavelets for an intensity analysis (von Tscharner, 2000). The wavelets are defined by the following formula: [(f/cf)^(cf*scale)]*[e^((-f/cf)+1)^(cf*scale)] where *f *is frequency of length *n*, *cf* is center frequency (defined below) and is an array of *j *columns and *n* row, and scale is a constant. cf = (1/scale)*(j

Hi Folks, I am dealing with data which have been presented as at each x_i, mean m_i of the y-values at x_i, sd s_i of the y-values at x_i number n_i of the y-values at x_i and I want to linearly regress y on x. There does not seem to be an option to 'lm' which can deal with such data directly, though the regression problem could be algebraically

Hi, I am running simulations that does multiple comparisons to control. For each simulation, I need to model 7 nls functions. I loop over 7 to do the nls using try if try fails, I break out of that loop, and go to next simulation. I get warnings on nls failures, but the simulation continues to run, except when the internal call (internal to nls) of the chol2inv fails.

Hi all, I am trying to concatenate words together to create new column names, using a loop. Please consider the following toy example: x <- matrix(nrow = 1, ncol = 3) colnames(x) <- c("a", "b", "c") x[1,1] <- "1" x[1,2] <- "2" x[1,3] <- "3" I would like to create a new matrix with column names based on the column names

Buenas tardes, Estoy interesado en generar observaciones de una distribucion binomial bivariada en la que hay _cierto_ grado de correlacion (denotemoslo rho). Podria por favor alguien indicarme como hacerlo en R? Este es el contexto. Supongamos que se tienen dos experimentos en los que la variable respuesta _sigue_ una distribucion binomial, i.e., X_i ~Binomial(n_i, p_i), i=1,2 y que, por ahora,

Dear R users, I?m a graduate students and in my master thesis I must obtain the values of the parameters x_i which maximize this Multinomial log?likelihood function log(n!)-sum_{i=1]^4 log(n_i!)+sum_ {i=1}^4 n_i log(x_i) under the following constraints: a) sum_i x_i=1, x_i>=0, b) x_1<=x_2+x_3+x_4 c)x_2<=x_3+x_4 I have been using the ?ConstrOptim? R-function with the instructions

Hi, I don't know what I am doing wrong to the toupper does not seem working in sub + regex. The following returns 's' not the upper class 'S' as I expect: sub("q_([a-z])[a-zA-Z]*",toupper('\\1'),"q_sviRaw") Can someone tell me where I did wrong? Thanks, Richard [[alternative HTML version deleted]]

Dear R users, I am writing to you because I have a few question on how to fix the error term variances in lme in the hope that you could help me. To my knowledge, the closest possibility is to fix the var-cov structure, but not the whole var-cov matrix. I found an old thread (a few years ago) about this, and it seems that the only alternative is to write the likelihood down and use optim or a

Dear helpers in this forum, This is a clarified version of my previous questions in this forum. I really need your generous help on this issue. > Suppose I have the following data set: > > id x y > 023 1 2 > 023 2 5 > 023 4 6 > 023 5 7 > 412 2 5 > 412 3 4 > 412 4 6 > 412 7 9 > 220 5 7 > 220 4 8 > 220 9 8 > ...... > Now I want to compute the

Hi, I would like to connect my GHA-system to R and to deliver the library to users of R. could you give me a comprehensive example containing the step by step procedure to: 1. create a callable library with R binaries (not having a standalone version of R with its own main()). If possible, I would like to run the system on unix,alpha,linux computers as well as on IBM's parallel

Hi, I sent the below message to r-devel but received no answer. I would like to contribute my system but I need som advice concerning the connections. Is my proposal of any interest to you? regards Ralf ?stermark ----- Forwarded message from Ralf ?stermark <rosterma at abo.fi> ----- Date: Mon, 21 Feb 2005 11:35:21 +0200 From: Ralf ?stermark <rosterma at abo.fi> Reply-To: Ralf

I have been playing around with sparse matrices in the Matrix package, in particularly with the Cholesky factorization of matrices of class dsCMatrix. And BTW, what a fantastic package. My problem is that I have to carry out repeated Cholesky factorization of a spares symmetric matrices, say Q_1, Q_2, ...,Q_n, where the Q's have the same non-zero pattern. I know in this case one does

Can someone clarify what I cannot make out from the documentation? The function 'clogit' in the 'survival' package is described as performing a "conditional logistic regression". Its return value is stated to be "an object of class clogit which is a wrapper for a coxph object." This suggests that its usefulness is confined to the sort of data which arise in

Dear All, In the past I have often used minpack (http://bit.ly/zXVls3) relying on the Levenberg-Marquardt algorithm to perform non-linear fittings. However, I have always dealt with a function of a single variable. Is there any difference if the function depends on two variables? To fix the ideas, please consider the function f(R,N)=(a/(log(2*N))+b)*R+c*N^d, where a,b,c,d are fit parameters. For