Simulation of an ARMA process

2 views (last 30 days)
ALBERT
ALBERT on 23 Oct 2015
Dear community,
As far as I know an ARMA process is that whose statistics changes over time, thus it is a non stationary process.
When i try to simulate an ARMA model I see that the mean does not change in time. I wonder why?
Here is the code: % --------------------------- N=1000; modSim = arima('Constant',0.2,'AR',{0.75,-0.4 0.32 -0.15}, 'MA',{0.7 -0.2 -0.5},'Variance',0.1); % ARMA(4,3) rng('default') Y = simulate(modSim,N, 'NumPaths', 1000);
figure plot(Y,'Color',[.85,.85,.85]) hold on h=plot(mean(Y,2),'k','LineWidth',2); legend(h,'Simulation Mean','Location','NorthWest') xlim([0,N]) title(strcat('Simulated ARMA(', num2str(length(modSim.AR)), ',', num2str(length(modSim.MA)), ') Series'));
% -----------------
Many thanks!!
Albert

Answers (0)

Categories

Find more on Conditional Mean Models in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!