The recently proposed Double Asymmetric GARCH-MIDAS (DAGM) model aims at separating the positive and negative macro variable variations within the long-run term and adds an asymmetric effect in the short-run component. In this work, the intent is to further extend the model in two main directions. A realized measure is included as a daily lagged variable in the short-run component (the socalled “–X” term) and a multi-step-ahead forecasting procedure is implemented for the class of GARCH–MIDAS (GM) models with the additional “–X” term. The extended DAGM-X model, which nests the DAGM and GM, is extensively evaluated under alternative configurations concerning the S&P 500 Index.
Il presente lavoro illustra una estensione del modello Double Asymmetric GARCH–MIDAS (DAGM), recentemente proposto. Nella modellizazione, oltre agli effetti asimmetrici nelle componenti di lungo e di breve periodo, `e stata introdotta una misura di volatilit`a realizzata giornaliera come variabile addizionale per la componente di breve periodo (la cosiddetta parte “–X”). Inoltre, `e stata sviluppata una procedura per le previsioni multi-step-ahead, valida per tutti i modelli GARCH– MIDAS (GM), anche con un termine aggiuntivo “–X”. La performance del DAGM– X, che generalizza il modello DAGM e il modello GM, `e stata valutata in riferimento all’indice S&P 500.
Double Asymmetric GARCH-MIDAS model: new insights and results / Amendola, Alessandra; Candila, Vincenzo; Maria Gallo, Giampiero. - (2020), pp. 927-932. (Intervento presentato al convegno SIS 2020: 50th Scientific meeting of the Italian Statistical Society tenutosi a PISA).
Double Asymmetric GARCH-MIDAS model: new insights and results
Alessandra Amendola
;Vincenzo Candila;
2020
Abstract
The recently proposed Double Asymmetric GARCH-MIDAS (DAGM) model aims at separating the positive and negative macro variable variations within the long-run term and adds an asymmetric effect in the short-run component. In this work, the intent is to further extend the model in two main directions. A realized measure is included as a daily lagged variable in the short-run component (the socalled “–X” term) and a multi-step-ahead forecasting procedure is implemented for the class of GARCH–MIDAS (GM) models with the additional “–X” term. The extended DAGM-X model, which nests the DAGM and GM, is extensively evaluated under alternative configurations concerning the S&P 500 Index.File | Dimensione | Formato | |
---|---|---|---|
Candila_Double-Asymmetric-SIS_2020.pdf
accesso aperto
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
421.18 kB
Formato
Adobe PDF
|
421.18 kB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.