Fisher Information

Description

fisherI utilizes nint_integrate to evaluate the Fisher information.

Usage

1
fisherI(ff, theta, parNames, yspace, ...)

Arguments

ff

either

  • function(y, theta, i, j, ...) which evaluates to the inner part of the expectation integral/sum.

  • list(f=function(y, theta, ...), d2logf=function(y, theta, i, j, ...)) (recommended)

  • list(f=function(y, theta, ...), dlogf=function(y, theta, i, ...))

where f is the joint probability density function and dlogf/d2logf is the first/second derivative of log(f) with respect to theta[[i]]/theta[[i]] and theta[[j]].

theta

the list of parameters.

parNames

a vector of names or indices, the subset of parameters to use.

yspace

a space, the support of y.

...

other arguments passed to ff.

Details

If ff is a list, it shall contain dlogf xor d2logf.

Value

fisherI returns a named matrix, the Fisher information.

See Also

buildf, numDerivLogf, DerivLogf, nint_space, nint_transform, nint_integrate, param

Examples

1
## see examples for param