similar to: multivariate quantile estimation

Hello, I have a question about multivariate Shapiro-Wilks test. I tried to analyze if the data I have are multivariate normal, or how far they are from being multivariate normal. However, any time I did >mshapiro.test(mydata) I get the message: Error in solve.default(R %*% t(R), tol = 1e-18) : system is computationally singular: reciprocal condition number = 5.38814e-021 I tried

Dear List, Is there any function in R to generate multivariate F distribution with given correlation/covariance matrix? Actually, I just want to generate some 2-dimentional non-normal data sets (skewed) for low (may be around 0.3 cor coeff.) negatively and also positively correlated variables ? Thanks in advance. Anna

Guys: 1).When I using the 'quantile' function, I get really confused. Here is what I met: > x<-zoo(rnorm(500,0,1)) > quantile(x,0.8) 400 1.060258 > c=rnorm(500,0,1) > quantile(c,0.8) 80% 0.9986075 why do the results display different? Is that because of the different type of the class? 2).And I want to use the 'rollapply' function to compute a

Dear R experts/ users Full Information Maximum Likelihood (FIML) estimation approach is considered robust over Seemingly Unrelated Regression (SUR) approach for analysing data of multivariate sample selection problem. The zero cases in my dependent variables are resulted from three sources: Irreverent options, not choosing due to negative utility and not used in the reported time. FIML can

Hello, I try to find a function for M-estimation in multivariate linear regression model (function that can estimate betas in my model: y=x * beta + e, where y is a matrix). I´ve searched R-site for a long time, but I am hopeless. I would like to ask, if there is any function for M-estimation in multivariate linear regression model in R. I know I can estimate betas in my model by rlm() function

Dear all, I would like to implement a spatial quantile regression using instrumental variable estimation (according to Su and Yang (2007), Instrumental variable quantile estimation of spatial autoregressive models, SMU economics & statistis working paper series, 2007, 05-2007, p.35 ). I am applying the hedonic pricing method on land transactions in Luxembourg. My original data set contains

I would like to estimate a 95% highest density area for a multivariate parameter space (In the context of anova). Unfortunately I have only experience with univariate kernel density estimation, which is remarkebly easier :) Using Gibbs, i have sampled from a posterior distirbution of an Anova model with k means (mu) and 1 common residual variance (s2). The means are independent of eachother, but

Mr./Ms. Thank your help I need the code of quantile regression - estimation of CAViaR, would do you like to help me! regards, tangyong school of managemnet ,fuzhou university, China [[alternative HTML version deleted]]

One of the real advances (in my humble oppinion of course) of 2003 is the Very Fast Multivariate Kernel Density Estimation algorithm by Alex Gray which achieves several order of speed improvement by using Computational Geometry to organize the data. The algorithm is now implemented in C++ with Mathlab interface by Alexander Ihler of MIT: http://ssg.mit.edu/~ihler/code/kde.shtml I wondered if a

Hi, I am trying to perform Maximum Likelihood estimation of a Multivariate model (2 independent variables + intercept) with autocorrelated errors of 1st order (ar(1)). Does R have a function for that? I could only find an univariate option (ar.mle function) and when writing my own I find that it is pretty memory-consuming (and sometimes wrong) so there must be a better way. Thanks, KB

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'm using the quantreg package for quantile regression. Although it's fine, there're is some weird behavior a little bit difficult to understant. In some occasions, the regression results table shows coefficients, t-statistics, standard errors and p-values. However, in other occasions it shows only coefficients and confidence intervals. Therefore, the question is... Is

Dear R-users, I've been comparing different types used in quantile() function and I found a thing I don't understand. When n*p[i] is less than 1, most types return the smallest vector element, while type 7 interpolates between the smallest and the second smallest element. Other types, that generally also use linear interpolation, don't do it. Example: > inp [1] -0.762498

Dear R users, A new package titled `np' is now available from CRAN. The package implements recently developed kernel methods that seamlessly handle the mix of continuous, unordered, and ordered factor datatypes often found in applied settings. The package also allows users to create their own nonparametric/semiparametric routines using high-level function calls (via the function npksum())

Dear R users, A new package titled `np' is now available from CRAN. The package implements recently developed kernel methods that seamlessly handle the mix of continuous, unordered, and ordered factor datatypes often found in applied settings. The package also allows users to create their own nonparametric/semiparametric routines using high-level function calls (via the function npksum())

(disclaimer: I'm in physics, not stats... ) I have a multivariate problem. One variable, call it R1, and 3 "predictor" variables, P1, P2, P3. My goal is to take a load of training data (I know R1,P1,P2,P3 for about 700 total points), and then predict R1 for a new set of data for which I have all the predictors. Simple, no? I understand how to calculate bandwidths, and I have a

I'd like to mention that there is a new quantile regression package "nprq" on CRAN for additive nonparametric quantile regression estimation. Models are structured similarly to the gss package of Gu and the mgcv package of Wood. Formulae like y ~ qss(z1) + qss(z2) + X are interpreted as a partially linear model in the covariates of X, with nonparametric components defined as

I am working on a two stage threshold quantile regression model in R, and my aim is to estimate the threshold of the reduced-form equation (call it rhohat), and the threshold of the structural equation (call it qhat), in two stages. On the first stage, i estimate rhohat by quantile regression and obtain the fitted values. I use these fitted values to estimate qhat on the second stage. The code is

Dear everyone, i`m a german economics student, writing my master´s thesis about "Multivariate Volatility Models". After having read about theoretical aspects of Multivariate GARCH ans Stochastic Volatility Models, I would like to compare DCC-GARCH and DC-SV with help of an empirical application. I figuered out that one has to use MCMC-simulation-methods for that. Some days ago I