ssa (version 1.3.0)

uni.npmle: Univariate NPMLE

Description

General nonparametric maximum likelihood estimation for a univariate mixing distribution, implemented using EM. Assumes that the observed data are \(X_i\) with marginal likelihood $$\int f(X_i;u)dG(u),$$ where \(G\) is the mixing distribution to be estimated. Suppose there are p observations and \(G\) is to be estimated on a grid of d points.

Usage

uni.npmle(D, maxit = 200, tol = 1e-04, verbose = FALSE)

Arguments

D

p x d matrix of conditional density values, where the \(ij\)th entry is \(f(X_i;u_j)\)

maxit

maximum number of EM iterations

tol

error tolerance

verbose

TRUE to print the error attained by each EM iteration

Value

g

d x 1 vector of probability masses at each grid point

Examples

# NOT RUN {
## generate parameters from mixing distribution
p <- 1000;
set.seed(1); theta <- rnorm(p);
## generate observed variables
X <- rnorm(p,theta,1);
## set grid points
d <- 25;
Theta <- seq(min(X),max(X),length=d);
## calculate D matrix
D <- outer(X,Theta,function(x,y){
  dnorm(x,y,1);
});
## fit npmle
g <- uni.npmle(D);
plot(Theta,g/sum(g*(Theta[2]-Theta[1])),type="l");
lines(Theta,dnorm(Theta),col="red");

# }