A1075
Title: Consistent order selection for ARFIMA processes
Authors: Kun Chen - Southwestern University of Finance and Economics (China) [presenting]
Ngai Hang Chan - City University of Hong Kong (Hong Kong)
Ching-Kang Ing - National Tsing Hua University (Taiwan)
Hsueh-Han Huang - Academia Sinica (Taiwan)
Abstract: Estimating the orders of the autoregressive fractionally integrated moving average (ARFIMA) model has been a long-standing problem in time series analysis. This challenge is tackled by establishing the Bayesian information criterion (BIC) consistency for ARFIMA models with independent errors. Since the memory parameter of the model can be any real number, this consistency result is valid for short memory, long memory and nonstationary time series. Further, the consistency of the BIC is extended to ARFIMA models with conditional heteroscedastic errors, thereby extending its applications to encompass many real-life situations. Finite-sample implications of the theoretical results are illustrated via numerical examples.
