Markov-Switching Dynamic Regression Models
Discrete-time Markov model containing switching state and dynamic
regression submodels
A Markov-switching dynamic regression model
describes the dynamic behavior of time series variables in the presence
of structural breaks or regime changes. A discrete-time Markov chain
(dtmc
) represents the discrete state space of the regimes
and specifies the probabilistic switching mechanism among the regimes. A
collection of dynamic regression (ARX or VARX) submodels (arima
or varm
) describes the dynamic
behavior of the time series within the regimes.
To create a Markov-switching dynamic regression model, see msVAR
.
Functions
Topics
Create Model
- Creating Markov-Switching Dynamic Regression Models
Learn requirements for creating a Markov-switching dynamic regression model by usingmsVAR
. - Create Univariate Markov-Switching Dynamic Regression Models
Create a fully or partially specified univariate Markov-switching dynamic regression model by usingmsVAR
. - Modify msVAR Model Specifications
Adjust the specifications of a created Markov-switching dynamic regression model. - Create Multivariate Markov-Switching Dynamic Regression Models
Create a fully or partially specified Markov-switching dynamic regression model for a multivariate response process by usingmsVAR
.
Fit Model to Data
- Analyze US Unemployment Rate Using Markov-Switching Model
Fit a univariate Markov-switching dynamic regression model of the US unemployment rate to time series data and simulate and forecast unemployment rate paths from the estimated model.
Generate Monte Carlo Simulations
- Simulate Univariate Markov-Switching Dynamic Regression Model
Generate random response and state paths from a two-state Markov-switching dynamic regression model. - Simulate Multivariate Markov-Switching Dynamic Regression Model
Generate random response and state paths from a three-state Markov-switching dynamic regression model. - Monte Carlo Simulation of Markov-Switching Dynamic Regression Model Response Variables
Characterize the distribution of a multivariate response series, modeled by a Markov-switching dynamic regression model, by summarizing the draws of a Monte Carlo simulation.