Within-regime volatility dynamics for observable- and Markov-switching score-driven models

We study the novel Markov-switching (MS) Beta-t-EGARCH (exponential generalized autoregressive conditional heteroscedasticity) model, using within-regime volatility dynamics, similar to the recent observable-switching (OS) Beta-t-EGARCH model. We report in-sample results on the Standard & Poor’s...

Full description

Saved in:

Bibliographic Details
Published in:	Finance research letters Vol. 73; p. 106631
Main Authors:	Blazsek, Szabolcs, Kong, Dejun, Shadoff, Samantha R.
Format:	Journal Article
Language:	English
Published:	Elsevier Inc 01.03.2025
Subjects:	Dynamic conditional score (DCS) Generalized autoregressive score (GAS) Regime-switching volatility models Standard & Poor’s 500 (S&P 500) C53 C52 C22 Generalized autoregressive score (GAS) Standard & Poor’s 500 (S&P 500) Dynamic conditional score (DCS) Regime-switching volatility models
ISSN:	1544-6123
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	We study the novel Markov-switching (MS) Beta-t-EGARCH (exponential generalized autoregressive conditional heteroscedasticity) model, using within-regime volatility dynamics, similar to the recent observable-switching (OS) Beta-t-EGARCH model. We report in-sample results on the Standard & Poor’s 500 (S&P 500) and a random sample of 50 firms from the S&P 500 from March 1986 to July 2024. We compare the out-of-sample forecasting performances of OS-Beta-t-EGARCH and MS-Beta-t-EGARCH from May 2005 to July 2024 and confirm that OS-Beta-t-EGARCH is superior to MS-Beta-t-EGARCH. •The performances of observable- and Markov-switching volatility models are compared.•All econometric models in this paper are score-driven times series models.•We use data on the S&P 500 index and 50 randomly selected stocks from the S&P 500.•Observable-switching is superior to Markov-switching both in- and out-of-sample.
ISSN:	1544-6123
DOI:	10.1016/j.frl.2024.106631