logliknormLR: Fast Computation of Marginal Predictive Loglikelihood

In PWD: Time Series Regression Using the Power Weighted Densities (PWD) Approach

Description

Helper function which takes as input a vector-valued response, y, a predictor matrix, X, a particular value of alpha, and init. It returns as an output the marginal predictive loglikelihood of the data given that value of alpha.

Usage

logliknormLR(yy, XX_aug, alpha, init)

Arguments

yy	T-length time series vector. y[1] represents the beginning of the time eries.
XX_aug	[T x (p+1)] dimensional matrix of covariates. This will include the intercept column.
alpha	PWD parameter we are calculating the marginal predictive loglikelihood for.
init	integer representing the time point to begin computing marginal predictive loglikelihood.

Value

Return a scalar value representing the marginal predictive loglikelihood of the data given alpha.

PWD documentation built on May 2, 2019, 10:21 a.m.