HermiteIntegration: Gauss-Hermite quadrature

In johannabertl/ApproxML: Approximate maximum likelihood estimation

Description

The one-dimensional integral of a function is numerically approximated with Gauss-Hermite quadrature with 20 nodes. See Abramowitz and Stegun 25.4.46 pp 890 and Table 25.10 pp 924.

Usage

HermiteIntegration(FUN, sig2 = 1, ...)

Arguments

FUN	function to be integrated
sig2	scalar
...	Additional arguments to the function given in FUN

Details

Integrals of the form int f(x) phi(x, sigma^2) dx are computed, where phi is the normal distribution density with mean 0 and variance sigma^2. FUN is the function f() and sig2 is sigma^2.

References

Abramowitz, M. & Stegun, I. A. (Eds.) Handbook of mathematical functions: with formulas, graphs, and mathematical tables. Dover Publications New York, 1972

johannabertl/ApproxML documentation built on May 22, 2019, 2:19 p.m.