Recent financial crises have placed an increased accent on methods dealing with risk management. Despite some critiques, the Value-at-Risk (VaR) still plays today a leading role among the risk measures. For this reason, the financial econometrics literature has been involved in proposing as much as possible accurate VaR models. Recently, the quantile regression (QR) approach has been used to directly forecast the VaR measures. Within such a QR framework, we add a (MI(xed)- DA(ta) Sampling) term to the well known Linear ARCH (LARCH) model. The MIDAS term allows the inclusion of macroeconomic variables usually observed at low frequencies (monthly, quarterly, and so forth) in contexts where the dependent variable is generally observed at higher frequencies (mainly, daily). The resulting model, named Quantile LARCH-MIDAS (Q–LARCH–MIDAS), is the first model incorporating the MIDAS approach within the QR framework.
Le recenti crisi finanziarie hanno portato un enorme interesse verso i metodi per la gestione del rischio. Nonostante alcune critiche, il Value-at-Risk (VaR) ha ancora oggi un ruolo primario tra le misure di rischio. Per questa ragione, la letteratura econometrica-finanziaria ha posto l’attenzione sui modelli per la stima del VaR. Recentemente, la regressione quantilica (QR) `e stata usata per calcolare direttamente il VaR. In questo contesto di QR, un termine MIDAS (MI(xed)-DA(ta) Sampling) `e aggiunto al noto modello Linear ARCH (LARCH). Il termine MIDAS permette l’inclusione di variabli macro, solitamente osservate a frequenza mensile o quadrimestrale, in contesti dove, di solito, la variabile dipendente `e osservata a cadenza giornaliera. Il modello risultante, chiamato Quantile LARCH-MIDAS (Q– LARCH–MIDAS), `e il primo modello che incorpora l’approccio MIDAS all’interno di un contesto di QR.
Adding MIDAS terms to Linear ARCH models in a Quantile Regression framework / Candila, Vincenzo; Petrella, Lea. - (2020), pp. 910-915. (Intervento presentato al convegno SIS 2020: 50th Scientific meeting of the Italian Statistical Society tenutosi a PISA).
Adding MIDAS terms to Linear ARCH models in a Quantile Regression framework
Vincenzo Candila
;Lea Petrella
2020
Abstract
Recent financial crises have placed an increased accent on methods dealing with risk management. Despite some critiques, the Value-at-Risk (VaR) still plays today a leading role among the risk measures. For this reason, the financial econometrics literature has been involved in proposing as much as possible accurate VaR models. Recently, the quantile regression (QR) approach has been used to directly forecast the VaR measures. Within such a QR framework, we add a (MI(xed)- DA(ta) Sampling) term to the well known Linear ARCH (LARCH) model. The MIDAS term allows the inclusion of macroeconomic variables usually observed at low frequencies (monthly, quarterly, and so forth) in contexts where the dependent variable is generally observed at higher frequencies (mainly, daily). The resulting model, named Quantile LARCH-MIDAS (Q–LARCH–MIDAS), is the first model incorporating the MIDAS approach within the QR framework.File | Dimensione | Formato | |
---|---|---|---|
Candila_Adding-MIDAS-SIS_2020.pdf
accesso aperto
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
417.87 kB
Formato
Adobe PDF
|
417.87 kB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.