klinfo: Kullback-Leibler Information

In TSSS: Time Series Analysis with State Space Model

Kullback-Leibler Information

Description

Compute Kullback-Leibler information.

Usage

klinfo(distg = 1, paramg = c(0, 1), distf = 1, paramf, xmax = 10)

Arguments

distg	function for the true density (1 or 2). 1 : Gaussian (normal) distribution paramg(1): mean paramg(2): variance 2 : Cauchy distribution paramg(1): \mu (location parameter) paramg(2): \tau^2 (dispersion parameter)
paramg	parameter vector of true density.
distf	function for the model density (1 or 2). 1 : Gaussian (normal) distribution paramf(1): mean paramf(2): variance 2 : Cauchy distribution paramf(1): \mu (location parameter) paramf(2): \tau^2 (dispersion parameter)
paramf	parameter vector of the model density.
xmax	upper limit of integration. lower limit xmin = -xmax.

Value

nint	number of function evaluation.
dx	delta.
KLI	Kullback-Leibler information, I(g;f).
gint	integration of g(y) over [-xmax, xmax].

References

Kitagawa, G. (2020) Introduction to Time Series Modeling with Applications in R. Chapman & Hall/CRC.

Examples

# g:Gauss, f:Gauss
klinfo(distg = 1, paramg = c(0, 1), distf = 1, paramf = c(0.1, 1.5), xmax = 8)

# g:Gauss, f:Cauchy
klinfo(distg = 1, paramg = c(0, 1), distf = 2, paramf = c(0, 1), xmax = 8)

TSSS documentation built on Sept. 29, 2023, 9:07 a.m.