Model Selection and Estimation of Long-Memory Time-Series Models

An exploratory estimation of ARFIMA(p,d,q) models showed that the estimated d is sensitive to the short-term dynamics included. To address this issue, I run a series of Monte Carlo experiments and test the performance (i) of the AIC and the SIC in selecting p and q and (ii) of the AIC, the SIC and the multimodel-inference approach of Burnham and Anderson (2002) in estimating d. I contribute to the literature by studying high-order data generating processes; by testing the MMI-approach; and by studying the impact of excluding models close to the data generating process from the set of candidate models. Three findings stand out. First, in terms of order selection, the SIC outperforms the AIC for low-order models but underperforms for high-order models. Second, the SIC still dominates both the AIC and the MMI-approach for inference. Third, set-up snooping has little impact.