STRUCTURAL BREAKS ESTIMATION FOR LONG MEMORY SIGNALS

We consider the problem of estimating the structural breaks in a long memory FARIMA process. The number m of break points as well as their locations, the order (p, d, q) and the parameters of each regime are assumed to be unknown. To estimate the unknown parameters, we propose two criteria based on the minimum description length (MDL) principle of Rissanen, namely a direct extension of MDL and an improved MDL criterion embedded with Bayes information criterion (BIC). A genetic algorithm is implemented to optimize these two criteria. Monte Carlo simulation results show that both criteria perform well for estimating the break points number and their locations. The direct extension of MDL tends to over-estimate the regimes model order which is not the case of the improved MDL criterion.