Abstract
Forecasting the risk of extreme losses is an important issue in the management of financial risk and has attracted a great deal of research attention. However, little attention has been paid to extreme losses in a higher frequency intraday setting. This paper proposes a novel marked point process model to capture extreme risk in intraday returns, taking into account a range of trading activity and liquidity measures. A novel approach is proposed for defining the threshold upon which extreme events are identified taking into account the diurnal patterns in intraday trading activity. It is found that models including covariates, mainly relating to trading intensity and spreads offer the best in-sample fit, and prediction of extreme risk, in particular at higher quantiles.
Similar content being viewed by others
Notes
Other functional forms of returns are possible to capture the intraday behavior, for instance, squared returns. However, according to preliminary results, the choice of absolute values is preferred.
We tried higher orders, but little is gained in the results observed.
The objective of this random variable \(\epsilon \) with small variance is to help to break ties between test values.
For instance, a risk manager could be more interested in the iid property of the exceptions during crisis period by choosing a lower level of a, instead of a correct unconditional coverage.
From a risk management perspective, using different values of \(\delta \) reflects different levels of risk-aversion. For instance, \(\delta =0.5\) gives a weight around \((0.6-\alpha )\) to the observations for which \(X_{t}<\hbox {VaR}_{\alpha }^{t}\), while that \(\delta =2.5\) and \(\delta =5\) give a weight around \((0.8-\alpha )\) and \((1-\alpha )\), respectively.
This is the check function introduced by Koenker and Bassett Jr (1978) and is defined as \(\ell \left( X_{t},\hbox {VaR}_{\alpha }^{t}\right) =\left( \mathbf {1}\left\{ X_{t}>\hbox {VaR}_{\alpha }^{t}\right\} -\alpha \right) \left( X_{t}-\hbox {VaR}_{\alpha }^{t}\right) \), where \(\mathbf {1}\left\{ X_{t}>\hbox {VaR}_{\alpha }^{t}\right\} \) is an indicator function that takes the value 1 when the return observed is larger than the VaR estimated for the day t at the given confidence level \(\alpha \) and the value 0 otherwise.
A Newey–West heteroskedasticity and autocorrelation-consistent (HAC) estimator is used to obtain the covariance matrix for these statistics.
References
Ahn H, Bae K, Chan K (2001) Limit orders, depth, and volatility: evidence from the stock exchange of Hong Kong. J Finance 56(2):767–788
Balkema AA, De Haan L (1974) Residual life time at great age. Ann Probab 2(5):792–804
Berger D, Chaboud A, Hjalmarsson E (2009) What drives volatility persistence in the foreign exchange market. J Financ Econ 94:192–213
Berkowitz J, Christoffersen P, Pelletier D (2011) Evaluating value-at-risk models with desk-level data. Manag Sci 57(12):2213–2227
Chavez-Demoulin V, McGill J (2012) High-frequency financial data modeling using Hawkes processes. J Bank Finance 36(12):3415–3426
Chavez-Demoulin V, Davison A, McNeil A (2005) A point process approach to value-at-risk estimation. Quant Finance 5:227–234
Chavez-Demoulin V, Embrechts P, Sardy S (2014) Extreme-quantile tracking for financial time series. J Econom 181(1):44–52
Chavez-Demoulin V, Embrechts P, Hofert M (2016) An extreme value approach for modeling operational risk losses depending on covariates. J Risk Insur 83:735–776. https://doi.org/10.1111/jori.12059
Chordia T, Roll R, Subrahmanyam A (2001) Market liquidity and trading activity. J Finance 56:501–530
Chordia T, Roll R, Subrahmanyam A (2002) Order imbalance, liquidity, and market returns. J Financ Econ 65:111–130
Clements A, Herrera R, Hurn A (2015) Modelling interregional links in electricity price spikes. Energy Econ 51:383–393
Diebold FX, Mariano RS (1995) Comparing predictive accuracy. J Bus Econ Stat 13(3):253–265
Embrechts P, Liniger T, Lin L et al (2011) Multivariate Hawkes processes: an application to financial data. J Appl Probab 48:367–378
Engle R, Manganelli S (2004) CAViaR. J Bus Econ Stat 22(4):367–381
Evans M, Lyons R (2002) Order flow and exchange rate dynamics. J Polit Econ 110:170–180
Giacomini R, White H (2006) Tests of conditional predictive ability. Econometrica 74(6):1545–1578
González-Rivera G, Lee TH, Mishra S (2004) Forecasting volatility: a reality check based on option pricing, utility function, value-at-risk, and predictive likelihood. Int J Forecast 20(4):629–645
Gresnigt F, Kole E, Franses P (2015) Interpreting financial market crashes as earthquakes: a new early warning system for medium term crashes. J Bank Finance 56:123–139
Groß-Klußmann A, Hautsch N (2011) When machines read the news: using automated text analytics to quantify high frequency news-implied market reactions. J Empir Finance 18:321–340
Hall A, Hautsch N (2006) Order aggressiveness and order book dynamics. Empir Econ 30:973–1005
Hall A, Hautsch N (2007) Modelling the buy and sell intensity in a limit order book market. J Financ Mark 10:249–286
Hasbrouck J, Seppi D (2001) Common factors in prices, order flows and liquidity. J Financ Econ 59:383–411
Herrera R, Gonzalez N (2014) The modeling and forecasting of extreme events in electricity spot markets. Int J Forecast 30(3):477–490
Herrera R, Schipp B (2013) Value at risk forecasts by extreme value models in a conditional duration framework. J Empir Finance 23:33–47
Herrera R, Schipp B (2014) Statistics of extreme events in risk management: the impact of the subprime and global financial crisis on the German stock market. N Am J Econ Finance 29:218–238
Koenker R, Bassett G Jr (1978) Regression quantiles. Econometrica 46:33–50
Koenker R, Xiao Z (2006) Quantile autoregression. J Am Stat Assoc 101(475):980–990
Kuester K, Mittnik S, Paolella MS (2006) Value-at-risk prediction: a comparison of alternative strategies. J Financ Econom 4(1):53–89
Kupiec PH (1995) Techniques for verifying the accuracy of risk measurement models. J Deriv 3(2):73–84
Liu S, Tse Y (2015) Intraday value-at-risk: an asymmetric autoregressive conditional duration approach. J Econom 189(2):437–446
McNeil A, Frey R (2000) Estimation of tail-related risk measures for heteroscedastic financial time series: an extreme value approach. J Empir Finance 7:271–300
Næs R, Skjeltorp JA (2006) Order book characteristics and the volume-volatility relation: empirical evidence from a limit order market. J Financ Mark 9:408–432
Ogata Y (1978) The asymptotic behaviour of maximum likelihood estimators for stationary point processes. Ann Inst Stat Math 30(1):243–261
Opschoor A, Taylor N, van der Wel M, van Dijk D (2014) Order flow and volatility: an empirical investigation. J Empir Finance 28:185–201
Pickands J III (1975) Statistical inference using extreme order statistics. Ann Stat 3:119–131
Reiss RD, Thomas M (2007) Statistical analysis of extreme values: with applications to insurance, finance, hydrology and other fields. Birkhäuser, Basel
Ziggel D, Berens T, Weiß GN, Wied D (2014) A new set of improved value-at-risk backtests. J Bank Finance 48:29–41
Acknowledgements
Herrera acknowledges the Chilean CONICYT funding agency for financial support (FONDECYT 1180672) for this project.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Herrera, R., Clements, A. A marked point process model for intraday financial returns: modeling extreme risk. Empir Econ 58, 1575–1601 (2020). https://doi.org/10.1007/s00181-018-1600-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00181-018-1600-y