Selective linear segmentation for detecting relevant parameter changes

Vous êtes ici

Accueil » Selective linear segmentation for detecting relevant parameter changes
30 Septembre 2019
Types de publication: 
Cahier de recherche
Auteur(s): 
Arnaud Dufays
Elysee Aristide Houndetoungan
Alain Coën
Axe de recherche: 
Enjeux économiques et financiers
Mots-clés: 
change-point
time-varying parameter
model selection
Hedge funds
Classification JEL: 
C11
C12
C22
C32
C52
C53

Change-point processes are one flexible approach to model long time series. We propose a method to uncover which model parameter truly vary when a change-point is detected. Given a set of breakpoints, we use a penalized likelihood approach to select the best set of parameters that changes over time and we prove that the penalty function leads to a consistent selection of the true model. Estimation is carried out via the deterministic annealing expectation-maximization algorithm. Our method accounts for model selection uncertainty and associates a probability to all the possible time-varying parameter specications. Monte Carlo simulations highlight that the method works well for many time series models including heteroskedastic processes. For a sample of 14 Hedge funds (HF) strategies, using an asset based style pricing model, we shed light on the promising ability of our method to detect the time-varying dynamics of risk exposures as well as to forecast HF returns.
 

Contact: 


Arnaud Dufays : Département des sciences de gestion, Université Namur et Département d'économique, Université Laval. CRREP et CeReFiM.
Elysee Aristide Houndetoungan : Département d'économique, Université Laval.
Alain Coën : Department of Finance, UQAM.