ISO: Class "ISO"

In thq80/Cornuet_2012_Adaptive-Mutiple-IS: A R Adaptive Multiple Importance Sampling

Description

An Importance Sampling Object

Objects from the Class

The suggested way to construct objects of class ISO is to use the constructor functions AIS(N, niter, p, target, initialize, mixture, verbose = FALSE, seed = NULL) and AMIS(N, niter, p, target, initialize, mixture, verbose = FALSE, seed = NULL)

Slots

IS:

A "matrix" the Importance Sample in the first p columns and the Importance Weights in the last column.

Prop:

A "numeric" with the mixture proportion for the maximum perplexity mixture.

Mean:

A "numeric" containing the mean of the importance sample.

Var:

A "matrix" containing the Variance Covariance Matrix of the importance sample.

Perp:

A "numeric" with the maximum perplexity.

ESS:

A "numeric", the Effective Sample Size.

envir:

A "environment", it contains functions with suffix r and p with the "usual" R meaning, for simulating from the target and for computing the value of the CDF. Their Argument lists are:

rTarg(n)

pTarg(q,lower.tail=TRUE,log.p=FALSE,N=1000)

seed:

A "integer", the initial seed for the pseudo-random number generator.

call:

The "call" to the function.

args:

A "list" with the arguments the function has been called with.

Methods

AIS(N,niter,p,target,proposal=mvtComp(df=3),initialize,mixture,verbose=FALSE,tol=0.001,seed=sample(.Random.seed,1))

A constructor function. See also AIS

AMIS(N,niter,p,target,proposal=mvtComp(df=3),initialize=uniInit(),mixture,verbose=FALSE,parallel = c("no", "multicore", "snow"),nCores=-1,cl=NULL,tol=0.001,seed=sample(.Random.seed,1),...)

A constructor function. See also AMIS

plot(x,whichOne=1L,prj=1L:2L,N=100,xlim=c(-30,30),ylim=c(-30,30),main,...)

signature(x = "ISO"): plotting method.

whichOne

Work in Progress

prj

For p>2 a vector of length 2 specifying the 2D marginal to be plotted.

N

Size of the sample to plot.

mean(x, fun=NULL,...)

signature(x = "ISO"): Returns the mean of fun(x).

var(x, y=NULL,na.rm=FALSE,use)

signature(x = "ISO"): Returns the Variance-Covariance of y(x).

weights

signature(object = "ISO"): Returns the IS weights.

sample

signature(x="ISO",size): To sample from the target distribution.

names

signature(x = "ISO"): alias for slotNames.

[[

signature(x = "ISO"): subsetting of the object.

$

signature(x = "ISO"): subletting of the object.

show

signature(object = "ISO"): the show method.

Author(s)

Luca Pozzi, [e-mail]

See Also

The package vignette demoARAMIS.

Examples

showClass("ISO")

thq80/Cornuet_2012_Adaptive-Mutiple-IS documentation built on May 21, 2019, 9:23 a.m.