similar to: gauss.quad.prob

Spencer Thanks for your thoughts on this. I did a bit of work and did end up with a method (more a trick), but it did work. I am certain there are better ways to do this, but here is how I resolved the issue. The integral I need to evaluate is \begin{equation} \frac{\int_c^{\infty} p(x|\theta)f(\theta)d\theta} {\int_{-\infty}^{\infty} p(x|\theta)f(\theta)d\theta} \end{equation} Where

I believe this may be more related to analysis than it is to R, per se. Suppose I have the following function that I wish to integrate: ff <- function(x) pnorm((x - m)/sigma) * dnorm(x, observed, sigma) Then, given the parameters: mu <- 300 sigma <- 50 m <- 250 target <- 200 sigma_i <- 50 I can use the function integrate as: > integrate(ff, lower= -Inf, upper=target)

Hi all, We know that Hermite polynomial is for Gaussian, Laguerre polynomial for Exponential distribution, Legendre polynomial for uniform distribution, Jacobi polynomial for Beta distribution. Does anyone know which kind of polynomial deals with the log-normal, Student抯 t, Inverse gamma and Fisher抯 F distribution? Thank you in advance! David [[alternative HTML version deleted]]

Hi, I would like to solve a double integral of the form \int_0^1 \int_0^1 x*y dx dy using Gauss Quadrature. I know that I can use R's integrate function to calculate it: integrate(function(y) { sapply(y, function(y) { integrate(function(x) x*y, 0, 1)$value }) }, 0, 1) but I would like to use Gauss Quadrature to do it. I have written the following code (using R's statmod package)

List: Thank you for the replies to my post yesterday. Gabor and Phil also gave useful replies on how to improve the function by relying on mapply rather than the explicit for loop. In general, I try and use the family of apply functions rather than the looping constructs such as for, while etc as a matter of practice. However, it seems the mapply function in this case is slower (in terms of CPU

Hi there, I am using SAS Proc NLMIXED to maximize a likelihood with multivariate normal random effects. An example is the two part random effects model for repeated measures semi-continous data with a cluster at 0. I use the "model y ~ general(loglike)" statement in Proc NLMIXED, so I can specify a general log likelihood function constructed by SAS programming statements. Then the

Does anyone see why my code does not integrate to 1? library(statmod) mu <- 0 s <- 1 Q <- 5 qq <- gauss.quad(Q, kind='hermite') sum((1/(s*sqrt(2*pi))) * exp(-((qq$nodes-mu)^2/(2*s^2))) * qq$weights) ### This does what's it is supposed to myNorm <- function(theta) (1/(s*sqrt(2*pi))) * exp(-((theta-mu)^2/(2*s^2))) integrate(myNorm, -Inf, Inf)

Hi all, Does anybody know any function that performs gaussian adapative quadrature integration of univariate functions? Thanks in advance, Regards, Caio __________________________________________________ [[alternative HTML version deleted]]

Can anyone help me on how to get the nodes and weights of the adaptive quadrature using R. Best wishes Boikanyo. ----- The University of Glasgow, charity number SC004401

Hi All, I am trying to use A Gaussian quadrature over the interval (-infty,infty) with weighting function W(x)=exp(-(x-mu)^2/sigma) to estimate an integral. Is there a way to do it in R? Is there a function already implemented which uses such weighting function. I have been searching in the statmode package and I found the function "gauss.quad(100, kind="hermite")" which uses

Dear R-users I would like to integrate something like \int_k^\infty (1 - F(x)) dx, where F(.) is a cumulative distribution function. As mentioned in the "integrate" help-page: integrate(dnorm,0,20000) ## fails on many systems. This does not happen for an adaptive Simpson or Lobatto quadrature (cf. Matlab). Even though I am hardly familiar with numerical integration the implementation

I am using both nlminb and optim to get MLEs from a likelihood function I have developed. AFAIK, the model I has not been previously used in this way and so I am struggling a bit to unit test my code since I don't have another data set to compare this kind of estimation to. The likelihood I have is (in tex below) \begin{equation} \label{eqn:marginal} L(\beta) = \prod_{s=1}^N \int

Dear list, I'm seeking some advice regarding a particular numerical integration I wish to perform. The integrand f takes two real arguments x and y and returns a vector of constant length N. The range of integration is [0, infty) for x and [a,b] (finite) for y. Since the integrand has values in R^N I did not find a built-in function to perform numerical quadrature, so I wrote my own after

I've just encountered a segfault when using DierckxSpline::percur function. Below is the minimal example which triggers the error: --- library(DierckxSpline) x <- 1:10 y <- rep(0, 10) pspline <- percur(x, y) --- *** caught segfault *** address (nil), cause 'memory not mapped' Traceback: 1: .Fortran("percur", iopt = as.integer(iopt), m = as.integer(m), x =

Dear list, [cross-posting from Stack Overflow where this question has remained unanswered for two weeks] I'd like to perform a numerical integration in one dimension, I = int_a^b f(x) dx where the integrand f: x in IR -> f(x) in IR^p is vector-valued. integrate() only allows scalar integrands, thus I would need to call it many (p=200 typically) times, which sounds suboptimal. The

I have constructed the function mml2 (below) based on the likelihood function described in the minimal latex I have pasted below for anyone who wants to look at it. This function finds parameter estimates for a basic Rasch (IRT) model. Using the function without the gradient, using either nlminb or optim returns the correct parameter estimates and, in the case of optim, the correct standard

Dear All, Can R perform multivariate integration with infinite limits of integration? Thanks in advance, Paul

Hi R Users, I'm going to estimate via. ML the parameters in Poisson Lognormal model. The model is: x | lambda ~ Poisson(lambda) lambda ~ Lognormal(a,b) Unfortunately, I haven't found a useful package allowing for such estimation. I tried to use "poilog" package, but there is no equations and it's hard to understand what exactly this package really does. Using it I get the

Dear All, I wonder if there is a way to fit a generalized linear mixed models (for repeated binomial data) via a direct Maximum Likelihood Approach. The "glmm" in the "repeated" package (Lindsey), the "glmmPQL" in the "MASS" package (Ripley) and "glmmGIBBS" (Myle and Calyton) are not using the full maximum likelihood as I understand. The

I'm just curious, but wondering if there has been any work in making these algorithms available in R. They are aimed at accelerating matrix-vector products using approximation ideas, and might be useful in applications such as kernel machines, Gaussian processes/kriging. Thanks Mark Palmer Landscape Monitoring and Modelling CSIRO Mathematical and