Testing for threshold effect in ARFIMA models: Application to US unemployment rate data

Lahiani, A.; Scaillet, Olivier

doi:10.1016/j.ijforecast.2009.01.004

Advanced search

Browse by...

Deposit

Highlights

More informations

Title		Testing for threshold effect in ARFIMA models: Application to US unemployment rate data
Authors		Lahiani, A. Scaillet, Olivier
Published in		International Journal of Forecasting. 2009, vol. 25, no. 2, p. 418-428
Abstract		Macroeconomic time series often involve a threshold effect in their ARMA representation, and exhibit long memory features. In this paper we introduce a new class of threshold ARFIMA models to account for this. The threshold effect is introduced in the autoregressive and/or fractional integration parameters, and can be tested for using LM tests. Monte Carlo experiments show the desirable finite sample size and the power of the test with an exact maximum likelihood estimator of the long memory parameter. Simulations also show that a model selection strategy is available to discriminate between the competing threshold ARFIMA models. The methodology is applied to US unemployment rate data, where we find a significant threshold effect in the ARFIMA representation, and a better forecasting performance relative to TAR and symmetric ARFIMA models.
Keywords		Threshold ARFIMA — LM test — Asymmetric time series
Identifiers		DOI: 10.1016/j.ijforecast.2009.01.004
Full text		Article (Accepted version) (195 Kb) - Free access Other version: http://linkinghub.elsevier.com/retrieve/pii/S0169207009000120
Structures		Faculté d'économie et de management
Research group		Geneva Finance Research Institute (GFRI)
Citation (ISO format)		LAHIANI, A., SCAILLET, Olivier. Testing for threshold effect in ARFIMA models: Application to US unemployment rate data. In: International Journal of Forecasting, 2009, vol. 25, n° 2, p. 418-428. doi: 10.1016/j.ijforecast.2009.01.004 https://archive-ouverte.unige.ch/unige:79876