Testing for mutually exciting jumps and financial flights in high frequency data

Dungey, Mardi; Erdemlioglu, Deniz; Matei, Marius; Yang, Xiye

doi:10.1016/j.jeconom.2017.09.002

Cited by 32 publications

(16 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A realized jump in the S& P 500 and VIX on 29 September 2008. Notes:Jumps in the S&P 500 and VIX index on September 29, 2008 (replicated from Dungey et al, 2018): 5‐minute intra‐day prices (left panels) and log‐returns (right panels). The circles on the right panels indicate a detected jump occurring simultaneously around 18:40 GMT…”

Section: Empirical Analysismentioning

confidence: 99%

Heterogeneous investment horizons, risk regimes, and realized jumps

Erdemlioglu

Gradojević

2020

Int. J Fin Econ

Self Cite

View full text Add to dashboard Cite

This paper introduces a new empirical framework to identify the regimes of jump‐type tail risk over multiple trading horizons. Our approach combines the hidden Markov regime‐switching model with realized jumps, which allows us to examine the tail risk exposure of investors at different investment scales. Applying our method to data on bonds, stocks, and currencies, we find evidence that market risk linked to jumps exhibits time‐varying regime shifts and horizon‐dependence. We show that high‐frequency trading does not contribute to market instability, as measured by jump risk. The European bond market appears to be more vulnerable to downside (left‐tail) jump risk, relative to the U.S. Treasury bond market. Market jump fear embedded in the VIX exhibits vertical clustering where both risk regimes at low frequencies remain unchanged at higher frequencies. These results suggest that the premium for jump risk not only depends on stress periods, but also on the frequency at which investors trade financial assets.

show abstract

Section: Empirical Analysismentioning

confidence: 99%

Heterogeneous investment horizons, risk regimes, and realized jumps

Erdemlioglu

Gradojević

2020

Int. J Fin Econ

Self Cite

View full text Add to dashboard Cite

show abstract

“…Second, standard nonparametric tests for jumps (e.g., Barndorff-Nielsen, Shephard and Winkel, 2006;Lee and Mykland, 2008;Podolskij and Vetter, 2009) aim to learn about the presence of jumps over a finite time span only. The same applies to Dungey, Erdemlioglub, Mateia and Yang's (2017) test for jump cross-excitation. However, the sequential implementation of these tests over rolling time spans induces severe size distortions.…”

Section: Introductionmentioning

confidence: 98%

“…1 All of the above papers restrict attention to the simultaneous arrival of jumps, and so where they originate is not an issue. Dungey et al (2017) take a different approach by studying jump transmission across assets/markets at the high frequency. To this end, they extend Boswijk, Laeven and Yang's (2018) test for jump self-excitation to a bivariate setting in order to test for jump cross-excitation in high frequency data.…”

Section: Introductionmentioning

confidence: 99%

Testing for Jump Spillovers Without Testing for Jumps

Corradi

Distaso²,

Fernandes³

2019

Journal of the American Statistical Association

View full text Add to dashboard Cite

This paper develops statistical tools for testing conditional independence among the jump components of the daily quadratic variation, which we estimate using intraday data. To avoid sequential bias distortion, we do not pretest for the presence of jumps. If the null is true, our test statistic based on daily integrated jumps weakly converges to a Gaussian random variable if both assets have jumps. If instead at least one asset has no jumps, then the statistic approaches zero in probability. We show how to compute asymptotically valid bootstrap-based critical values that result in a consistent test with asymptotic size equal to or smaller than the nominal size. Empirically, we study jump linkages between US futures and equity index markets.We find not only strong evidence of jump cross-excitation between the SPDR exchange-traded fund and E-mini futures on the S&P 500 index, but also that integrated jumps in the E-mini futures during the overnight period carry relevant information.

show abstract

“…The time-varying nature of asset correlations has long been of much interest to financial practitioners, with short-term increases in the co-movement of financial returns widely documented around periods of market stress (Preis et al, 2012;Packham & Woebbeking, 2019;Campbell et al, 2002;Cappiello et al, 2006;Billio et al, 2010). From a portfolio construction perspective, shortterm increases in asset correlations have frequently been linked to spikes in market volatility (Campbell et al, 2002) -with explanatory factors ranging from impactful news announcements (At-Sahalia & Xiu, 2016) to "riskon risk-off" effects (Dungey et al, 2018) -indicating that diversification can breakdown precisely when it is needed the most. In terms of systemic risk, increased co-movement between market returns is viewed as a sign of market fragility (Kritzman et al, 2011), as the tight coupling between markets can deepen drawdowns B. Lim, S. Zohren and S. Roberts are with the Department of Engineering Science and the Oxford-Man Institute of Quantitative Finance, University of Oxford, Oxford, United Kingdom (email:{blim, zohren, sjrob}@robots.ox.ac.uk). when they occur, resulting in financial contagion.…”

Section: Introductionmentioning

confidence: 99%

Detecting Changes in Asset Co-Movement Using the Autoencoder Reconstruction Ratio

2020

View full text Add to dashboard Cite

Detecting changes in asset co-movements is of much importance to financial practitioners, with numerous risk management benefits arising from the timely detection of breakdowns in historical correlations. In this article, we propose a real-time indicator to detect temporary increases in asset co-movements, the Autoencoder Reconstruction Ratio, which measures how well a basket of asset returns can be modelled using a lower-dimensional set of latent variables. The ARR uses a deep sparse denoising autoencoder to perform the dimensionality reduction on the returns vector, which replaces the PCA approach of the standard Absorption Ratio (Kritzman et al., 2011), and provides a better model for non-Gaussian returns. Through a systemic risk application on forecasting on the CRSP US Total Market Index, we show that lower ARR values coincide with higher volatility and larger drawdowns, indicating that increased asset co-movement does correspond with periods of market weakness. We also demonstrate that short-term (i.e. 5-min and 1-hour) predictors for realised volatility and market crashes can be improved by including additional ARR inputs.

show abstract

Testing for mutually exciting jumps and financial flights in high frequency data

Cited by 32 publications

References 82 publications

Heterogeneous investment horizons, risk regimes, and realized jumps

Heterogeneous investment horizons, risk regimes, and realized jumps

Testing for Jump Spillovers Without Testing for Jumps

Detecting Changes in Asset Co-Movement Using the Autoencoder Reconstruction Ratio

Contact Info

Product

Resources

About