ARFIMA simulations

Time series simulation with ARFIMA models.

Description

The code performs the simulation of time series with autoregressive fractionally integrated moving average (ARFIMA) models that generalize ARIMA (autoregressive integrated moving average) and ARMA autoregressive moving average models. ARFIMA models allow non-integer values of the differencing parameter and are useful in modeling time series with long memory. The code generally simulates an ARFIMA(p,d,q) model where d is the differencing parameter and p and q are the order of the autoregressive and moving average parts of the model respectively.

• Acknowledgements: This submission has inspired the following:; ARFIMA(p,d,q) estimator
• Required Products: Statistics Toolbox
• MATLAB release: MATLAB 7.5 (R2007b)

[György Inzelt]

Proved to be an invaluable tool!