randomizeInitState: Transformation of vector with initial states I for various...

In MMLR: Fitting Markov-Modulated Linear Regression Models

Description

Additional function to be used for simulation purposes (academical or research). Transforming of vector with initial states I for various observations with respect to stationary distribution of the states for the random environment.

Usage

randomizeInitState(StatPr, X, p = 1)

Arguments

StatPr	Vector (m x 1), m - number of states, m = 2,3,.. .The vector with stationary probabilities, user-defined vector.
X	Matrix (n x k), n - number of observations, k - number of columns (k - 1 - number of regressors). The matrix is needed to get the number of observations.
p	Scalar (from 1 to +inf), random number for simulation. The default value is 1.

Details

The initial states (m - number of states, m = 2,3,...) for various observations are independent and are chosen with respect to stationary distribution of the states for the random environment. The vector with stationary probabilities is user-defined vector.

Value

Vector with new initial states, according to stationary distribution of the states for the random environment.

Examples

Xtest <- cbind(rep_len(1,10),c(2,5,7,3,1,1,2,2,3,6), c(5,9,1,2,3,2,3,5,2,2))
StatPr <- matrix (c(0.364,0.242,0.394), nrow = 3, ncol = 1)
randomizeInitState(StatPr,Xtest,1)

MMLR documentation built on Jan. 9, 2020, 5:06 p.m.