Forecasting intraday market risk: A marked self-exciting point process with exogenous renewals

Tom Stindl

Journal of Empirical Finance, 2023, vol. 70, issue C, 182-198

Abstract: Methods to forecast intraday market risk are increasingly important in modern empirical finance due to the large volume of high-frequency trading. We propose a marked renewal Hawkes process to model the occurrence times of losses that exceed a threshold (exceedances) and a generalized Pareto distribution (GPD) enhanced with a time dependent scale parameter to model the size of the losses above the threshold (loss excesses). The scale parameter of the GPD evolves dynamically based on past exceedance occurrence times and loss excesses. A quantile autoregression model is used to define the exceedances as losses that exceed a time dependent threshold to accommodates for the cyclic trends of intraday trading. The arrival process of exogenous exceedances forms a renewal process and we investigate different waiting time distributions by applying the models to ASX stock data. We find evidence that the log-normal waiting time distribution provides the best quality in-sample fit among the competing models. The reliability of the forecasted market risk measures is assessed through backtesting which confirms the superior forecasting of intraday market risk by using our proposed strategy.

Keywords: Backtesting; Expected-shortfall; Renewal Hawkes; Value-at-risk; Quantile autoregression (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0927539822001104
Full text for ScienceDirect subscribers only

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:eee:empfin:v:70:y:2023:i:c:p:182-198

DOI: 10.1016/j.jempfin.2022.12.005

Access Statistics for this article

Journal of Empirical Finance is currently edited by R. T. Baillie, F. C. Palm, Th. J. Vermaelen and C. C. P. Wolff

More articles in Journal of Empirical Finance from Elsevier
Bibliographic data for series maintained by Catherine Liu ().