example_ms_ar_for.m

Autoregressive Markov Switching Model函数用于评估、仿真及预测自回归的马尔可夫转换模型。可以选择用于模型估计的分布函数。用于研究时间序列结构性变化

% Example Script for MS_AR_Fit.m (run it in the same directory)

clear;

addpath('m_Files');

load Example_Data.mat; % load .mat file

ar=4;                       % Number of lags in autoregressive component
k=2;                        % Number of states  
x=ret;                      % Time series from .mat file
advOpt.distrib='Normal';    % Distribution to use ('Normal' or 't' - default = 'Normal')
advOpt.std_method=1;        % Method for standard error calculation

[Spec_Output]=MS_AR_Fit(x,ar,k,advOpt);    % First estimate the model

[meanFor,stdFor]=MS_AR_For(Spec_Output,x,advOpt);  % call for forecasting procedure

fprintf(1,['\nThe mean forecast at t+1 is ',num2str(meanFor)]);
fprintf(1,['\nThe sigma forecast at t+1 is ',num2str(stdFor),'\n']);

rmpath('m_Files');