Un modello a soglia per la volatilità del mercato azionario italiano: performance previsive e valutazione del rischio di portafoglio

Asmara Jamaleh
Additional contact information
Asmara Jamaleh: Servizio Studi e Analisi Finanziaria Banca Commerciale Italiana, Milano

Rivista di Politica Economica, 2001, vol. 91, issue 2, 79-132

Abstract: A Self-Exciting Threshold AutoRegressive (SETAR) model is applied to the Italian stock market volatility, to obtain volatility forecasts and Value-at-Risk (VaR) estimates. There is almost nothing dealing with Italian markets in the literature of Threshold models, which have never been used for VaR purposes up to now. The SETAR model's performance is compared to competitive linear and GARCH specifications and to the JP Morgan's RiskMetrics™ method. Here, the SETAR model shows the best performance in predicting volatility and VaR values, thanks to its ability in capturing some major volatility's dynamics. Only the Threshold model is able to distinguish an extraordinary from a persistent market shock. Its superiority is more evident during critical market periods. As also for the tendency to over/underestimate VaR values, the SETAR model presents major advantages. This is crucial in investment decisions. The model selection is performed using Tsay's procedure, whose effectiveness is successfully tested here.

JEL-codes: C12 C22 C51 C52 C53 G10 G11 G14 G20 (search for similar items in EconPapers)
Date: 2001
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.rivistapoliticaeconomica.it/2001/febbraio/jamaleh.php
Payment required

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:rpo:ripoec:v:91:y:2001:i:2:p:79-132

Access Statistics for this article

Rivista di Politica Economica is currently edited by Gustavo Piga

More articles in Rivista di Politica Economica from SIPI Spa
Bibliographic data for series maintained by Sabrina Marino ().