similar to: Kernel density / weights matrix?

Greetings, I have the following? Problem: Given k (=10) discrete independent random variables X_i with n_i (= 5 to 20) values each,compute quantiles of the distribution of the sum X = X_1+...+X_k. Here X has n=n_1 x n_2 ... n_k distinct values which is too large to list them all together with their probabilities. I tried several approaches: (A) Convolution: each X_j is approximated with

Greetings. For R gurus this may be a no brainer, but I could not find pointers to efficient computation of this beast in past help files. Background - I wish to implement a Cramer-von Mises type test statistic which involves double sums of max(X_i,Y_j) where X and Y are vectors of differing length. I am currently using ifelse pointwise in a vector, but have a nagging suspicion that there is a

I discovered recently that the phrase "Oja median" produces no hits in Jonathan Baron's very valuable R search engine. I found this surprising since I've long regarded this idea as one of the more interesting notions in the multivariate robustness literature. To begin to remedy this oversight I wrote a bivariate version and then decided that writing a general p-variate version

Hello This may have been answered elsewhere, and I have looked on the web, but nothing helps. I am trying to do the following: X<-matrix(c(1:15),nrow=3,byrow=T) Y<-matrix(c(2,4,6,8,10),ncol=1) I need to sum the product of each row of X by the remaining j rows multiplied by j y values (i.e sum( t(x_i) x_j y_j) ) Hope this makes sense. Thanks in advance. Ps: how do I reference all

Hi all, I have been trying to add the line: h = n - p0 + 1; just after h = n / 2 + 1; (line 131) in the source code (the original paper mention this is possible for p0&lt;n). with p0 declared as an int (actually i used the same declaration protocol as n everywhere in the code). The &quot;new&quot; source compiles, but when i give it reasonable values of p0, it runs unto

This is a statistics question not an R question. When calculating the sample correlation coefficient cor(x_t,y_t) between say two variables, x_t and y_t t=1,.....n ( one can assume that the variables are in time but I don't think this really matters for the question ), does someone know where I can find any piece of literature that says that each (x_j,y_j) pair has To be independent from the

R Users, We would like to announce that Version 1.0.1 of the energy package is now available on CRAN. The energy package introduces a new class of statistical tests based on the concept of Newton's potential energy. Included in the package are * mvnorm.etest (test) and mvnorm.e (statistic) A rotation invariant multivariate goodness-of-fit test, implemented for testing

R Users, We would like to announce that Version 1.0.1 of the energy package is now available on CRAN. The energy package introduces a new class of statistical tests based on the concept of Newton's potential energy. Included in the package are * mvnorm.etest (test) and mvnorm.e (statistic) A rotation invariant multivariate goodness-of-fit test, implemented for testing

Hello r-help, I am using the excellent np package to conduct a nonparametric kernel regression and am having some trouble plotting the results. I have 2 covariates, x1 and x2, and a continuous outcome variable y. I am conducting a nonparametric regression of y on x1 and x2. The one somewhat unusual feature of these data is that, to be included in the dataset, x1 must be at least as large as x2.

Dear R-helpers, First I apologize if my question is quite simple I have a large datasets which more 100 variables. For a research I need to change all name of variables with add one or more letters on each variables. For example, > data(Pima.tr) > Pima.tr[1:5,] npreg glu bp skin bmi ped age type 1 5 86 68 28 30.2 0.364 24 No 2 7 195 70 33 25.1 0.163 55 Yes 3 5

Hello, I have the following function that receives a "function pointer" formal parameter name "fnc": loocv <- function(data, fnc) { n <- length(data.x) score <- 0 for (i in 1:n) { x_i <- data.x[-i] y_i <- data.y[-i] yhat <- fnc(x=x_i,y=y_i) score <- score + (y_i - yhat)^2 } score <- score/n

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

Dear all, I am using npreg from the np library to run a Kernel regression. My dataset is relatively large and has about 3000 observations. The dependent variable is continuous and I have a total of six independent variables -- two continuous, two ordinal and two categorical. The model converges without problems but it takes a very long time to do so (nearly one hour). Is there any way to speed

Dear all For my research, I am learning classification now. I was trying some example about classification tree pakages, such as tree and rpart, for instance, in Pima.te dataset have 8 variables (include class=type): library(rpart) library(datasets) pima.rpart <- rpart(type ~ npreg+glu+bp+skin+bmi+ped+age,data=Pima.te, method='class') plot(pima.rpart, uniform=TRUE) text(pima.rpart)

Hi All, The following is a snippet of my code. It works fine but it is very slow. Is it possible to speed it up by using different data structure or better solution? For 40000 runs, it takes 8 minutes now. Thanks a lot fun_activation = function(s_k, s_hat, x_k, s_hat_len) { common = intersect(s_k, s_hat); if(length(common) != 0) { index = match(common, s_k);

Dear List, I'm just teaching myself semi-parametric techniques. Apologies in advance for the long post. I've got observational data and a longitudinal, semi-parametric model that I want to fit in GAM (or potentially something equivalent), and I'm not sure how to do it. I'm posting this to ask whether it is possible to do what I want to do using "canned" commands

Dear R users, An updated version of the np package has recently been uploaded to CRAN (version 0.20-0). Version 0.20-0 is documented in Tristen Hayfield and Jeffrey S. Racine (2008). Nonparametric Econometrics: The np Package. Journal of Statistical Software 27(5). URL http://www.jstatsoft.org/v27/i05/ and also in a vignette (vignette("np",package="np")). There is also a FAQ

Dear R users, An updated version of the np package has recently been uploaded to CRAN (version 0.20-0). Version 0.20-0 is documented in Tristen Hayfield and Jeffrey S. Racine (2008). Nonparametric Econometrics: The np Package. Journal of Statistical Software 27(5). URL http://www.jstatsoft.org/v27/i05/ and also in a vignette (vignette("np",package="np")). There is also a FAQ

Greetings, Fellow R-ians: I'm working with a few different quasi-likelihood formulations for some data I'm analyzing. I'd like to implement the Information Matrix Test (see, e.g., White, 1982, or Lancaster, 1984) for each of them to determine which of the models is more likely. Since the null distribution of the test statistic is chi-squared, I envision calculating the ordinate of

I am searching for an R version of the code written in GAUSS by Bruce Hansen for his paper on Econometrica, 2000, "Sample Splitting and Threshold Estimation". Someone can help me? Davide -- Davide Fiaschi Dipartimento di Scienze Economiche University of Pisa Via Ridolfi 10 56100 Pisa (PI) Italy Phone/Fax: ++39.050.2216208/++39.050.598040 E-mail: dfiaschi at ec.unipi.it Homepage: