Tail Risk Research Articles

Countercyclical Income Risk and Portfolio Choices: Evidence from Sweden

ABSTRACTUsing Swedish administrative panel data, we document that workers facing higher left‐tail income risk when equity markets perform poorly have lower portfolio equity share. In line with theory, the relationship between cyclical skewness and stock holdings increases with the share of human capital in a worker's total wealth and vanishes as workers get closer to retirement. Cyclical skewness also predicts portfolio differences within pairs of identical twins. Our findings show that households hedge against correlated tail risks, an important mechanism in asset pricing and portfolio choice models.

The tail risk safe haven property of China's energy futures against US market implied volatility

This paper analyses the tail risk contagion of US market implied volatility (USIV) on China's energy futures (CEF) markets, exploring how to utilize operations in the CEF to achieve a safe haven. Leveraging CEF characteristics to simultaneously take both long-/short-positions and engage in long-/short-run investment horizons, this paper defines eight different CEF safe haven attributes to counteract the tail risk of extreme increases in USIV. Using trading data from March 27, 2018, to October 30, 2023, the empirical results show that, first, in the analysis of the entire sample period, China's coking coal futures can serve as a weak safe haven, aiding long-position investors in mitigating the tail risks associated with US gold and stock market implied volatility. Coking coal futures also assist short-position investors in countering US stock market implied volatility tail risk. Second, in the sub-period analysis, the safe haven attributes of CEF exhibit strong heterogeneity and asymmetry across different periods. Finally, the time span during which CEF exhibits a particular safe haven attribute does not persist for an extended period.

Managerial structure in the hedge fund industry

This paper provides the first study on how management structure influences hedge fund performance and risk. We document that hedge funds less tied to traditional assets often choose solo management structures. Solo-managed funds outperform team-managed funds, exhibit better skills in market return, volatility, and crisis timing, and demonstrate greater activity in beta management, but have higher idiosyncratic and tail risks. They are also less likely to be liquidated, with fund flows less performance sensitive. Using a sample of switched funds, we find that fund performance, assets, and risk correlate with the management structure switching decision.

The impact of the tail risk of demand on corporate investment: Evidence from Chinese manufacturing firms

The tail risk of demand over the past certain periods not only affect entrepreneurs' expectations on future demand uncertainty, increase the option value of “waiting an see” for delay investment, but also affect entrepreneurs' confidence and risk appetite by adjusting the worst-case scenario in the “demand belief set”, thereby further inhibiting investment. Utilizing the year-on-year sales growth data of various manufacturing industries in China, and the General Pareto Distribution method of Extreme Value Theory to estimate tail-index parameter and tail threshold as proxies for the tail risk of industry-level demand. Combined with the firm level data, our empirical results show that the tail risk indeed affect firm investment negatively, especially after the outbreak of global financial crisis, but such focal effect is stronger for firms with non-state ownership, small size, lower productivity and stricter financing constraints. Expansionary macroeconomic policies such as monetary and fiscal policy in periods of high tail risk have a higher incentive effect on firm investment, which provides more evidence for the effectiveness of macroeconomic policies during the economic recessions.

Volatility Spillover in the Turkish Financial Market: A QVAR Analysis

Purpose –The volatility spillover is crucial matter for policy makers and portfolio managers to understand risk transmission between financial markets to understand where potential loss and risk comes from. In this research, it is aimed to investigate the tail risk spilloveramong the BIST-100 stock index, TR 10-year bonds, USD-TL exchange rate, gold futures, and Brent petroleum in Türkiye.Design/Methodology/Approach –The quantile vector autoregressive(QVAR) model, recommended by Ando et al., (2022),is used in this study. The QVAR model is proposed method in the literature to capture the tail risk spillovers from very low to very high volatility in financial markets. The data is weekly frequency and spans from January 28, 2010, to December 8, 2023. The weekly volatility data is obtained from a formula that utilize daily maximum and minimum prices as described in Diebold and Yilmaz (2012). Findings –The output of this study indicates that the volatility spilloversbetween related markets differs across different quantiles. Other results reveal that stock, bond, and currency markets are net risk spillovers during extremely low and moderately volatile periods, but gold and oil are net risk receivers. However, USD/TL is the only risk-transmitter in times of excessive volatility. Moreover, the time-varying spillover analysis shows that the total spillover index hit records during the COVID-19 outbreaks.Discussion –The output of this study confirms the findings of previous studies that find the spillover index does not remain constant over different quantiles. The output of this study provides crucial insights to finance authorities and investors on the nature of market risk and strategies for its management.

The impact of climate policy on the risk contagion in China’s stock markets

ABSTRACT This paper analyses the impact of climate policy on the risk transmissions in China’s stock markets from sectoral perspectives. By constructing a (Generalized Autoregressive Conditional Heteroskedasticity-Mixed Data Sampling) GARCH-MIDAS model and conducting quantile regression tests, we find that climate policy exacerbates the tail risk spillover effect within the stock markets, while the sectoral effects would present heterogeneity. Both the spillovers and the systemic importance of finance and energy-intensive subsectors are susceptible to climate policy shocks. As for agriculture, its cross-sector spillovers are considerably intensified by climate policy, though its systemic importance is only enhanced by the policy in the long-run.

Risk contagion and diversification among sovereign CDS, stock, foreign exchange and commodity markets: Fresh evidence from G7 and BRICS countries

This paper investigates tail risk contagion and diversification effects among sovereign credit default swap (SCDS), foreign exchange, stock, and commodity markets by constructing a tail event driven network. Network adjacency matrices indicate more significant cross-border risk contagion effects compared to cross-market contagion. Besides, the systemic risk decomposition results demonstrate that SCDS markets play a dominant role in the aggregate risk level, while the foreign exchange markets effectively reduce systemic risk. Moreover, tail event driven network quantile regression analysis shows that the impact of network factors at different risk levels exhibits a coexistence of symmetry and asymmetry, accompanied by market heterogeneity.

PointCVaR: Risk-Optimized Outlier Removal for Robust 3D Point Cloud Classification

With the growth of 3D sensing technology, the deep learning system for 3D point clouds has become increasingly important, especially in applications such as autonomous vehicles where safety is a primary concern. However, there are growing concerns about the reliability of these systems when they encounter noisy point clouds, either occurring naturally or introduced with malicious intent. This paper highlights the challenges of point cloud classification posed by various forms of noise, from simple background noise to malicious adversarial/backdoor attacks that can intentionally skew model predictions. While there's an urgent need for optimized point cloud denoising, current point outlier removal approaches, an essential step for denoising, rely heavily on handcrafted strategies and are not adapted for higher-level tasks, such as classification. To address this issue, we introduce an innovative point outlier cleansing method that harnesses the power of downstream classification models. Using gradient-based attribution analysis, we define a novel concept: point risk. Drawing inspiration from tail risk minimization in finance, we recast the outlier removal process as an optimization problem, named PointCVaR. Extensive experiments show that our proposed technique not only robustly filters diverse point cloud outliers but also consistently and significantly enhances existing robust methods for point cloud classification. A notable feature of our approach is its effectiveness in defending against the latest threat of backdoor attacks in point clouds.

Research on Tail Risk Hedging in the Digital Asset Market

In recent years, the digital asset market has experienced rapid growth, but at the same time, its implicit tail risks have gradually raised concerns, especially amidst the global economic downturn caused by the pandemic. This paper aims to study tail risks and propose hedging strategies. Based on the GARCH and VaR models, this paper uses weekly return data from July 2017 to December 2021 for Bitcoin digital cryptocurrency and traditional non-digital asset market indices such as stocks, bonds, gold, and commodities. Utilizing EViews and Excel, the study explores tail risk measurement and portfolio diversification effects in the digital asset market. The main findings are as follows: (1) During the sample period, cryptocurrency returns exhibited significant volatility, with weak correlations with returns from traditional non-digital assets. Further analysis revealed strong hedging effects of digital assets against the bond market but weaker effects against the gold and stock markets, as well as commodity markets; (2) Cryptocurrency in traditional investment portfolios can enhance the risk-return ratio by adding it to the asset mix.

Predicting tail risks by a Markov switching MGARCH model with varying copula regimes

AbstractTo improve the dynamic assessment of risks of speculative assets, we apply a Markov switching MGARCH approach to portfolio risk forecasting. More specifically, we take advantage of the flexible Markov switching copula multivariate GARCH (MS‐C‐MGARCH) model of Fülle and Herwartz (2022). As an empirical illustration, we take the perspective of a risk‐averse agent and employ the suggested model for assessments of future risks of portfolios composed of a high‐yield equity index (S&P 500) and two safe‐haven investment instruments (i.e., Gold and US Treasury Bond Futures). We follow recent suggestions to employ the expected shortfall as a prime assessment of tail risks. To accurately evaluate the merits of the new model, we back‐test the risk forecasting for daily returns over 10 years for heterogeneous market environments including, for example, the COVID‐19 pandemic. We find that the MS‐C‐MGARCH model outperforms benchmark volatility models (MGARCH, C‐MGARCH) in predicting both value‐at‐risk and expected shortfall. The superiority of the MS‐C‐MGARCH model becomes stronger, when the share of comparably risky assets in the portfolio is relatively large.

Statistical Inference and Application of Asymmetrical Generalized Pareto Distribution Based on Peaks-Over-Threshold Model

Generalized Pareto distribution (GPD), an asymmetrical distribution, primarily models exceedances over a high threshold in many applications. Within the peaks-over-threshold (POT) framework, we consider a new GPD parameter estimation method to estimate a common tail risk measure, the value at risk (VaR). The proposed method is more suitable for the POT framework and makes full use of data information. Specifically, our estimation method builds upon the generalized probability weighted moments method and integrates it with the nonlinear weighted least squares method. We use exceedances for the GPD, minimizing the sum of squared differences between the sample and population moments of a function of GPD random variables. At the same time, the proposed estimator uses three iterations and assigns weight to further improving the estimated performance. Under Monte Carlo simulations and with a real heavy-tailed dataset, the simulation results show the advantage of the newly proposed estimator, particularly when VaRs are at high confidence levels. In addition, by simulating other heavy-tailed distributions, our method still exhibits good performance in estimating misjudgment distributions.

Green finance under stress: Unraveling the spillover effects of tail risk

The study was undertaken in direct response to escalating concerns regarding the environmental implications of green assets, with the primary objective of determining the potential of green financial markets to serve as a mitigating force against the risks associated with green stocks, bonds, cryptocurrencies, and carbon markets. To evaluate the risk, the study employed tail risk at the 5% value-at-risk level using the conditional autoregressive value-at-risk (CAViaR) and time-varying parameters vector auto regressions (TVP-VAR) techniques. From the analysis it was found that there was a strong intra-class connectedness between the assets but limited interconnectedness between the markets. Furthermore, we conducted an analysis of dynamic NET spillovers to discern the markets responsible for transmitting and receiving risk spillovers across diverse markets, with the aim of comprehending their behavior both before and after the onset of the COVID-19 pandemic. It was also found that extreme risk spillovers were observed during the study period. These findings have important implications for policymakers, regulators, investors, and other financial market participants. Green assets are designed to combat climate risk by promoting sustainable energy practices and reducing carbon emissions and to maintain the stability of the financial markets under both extreme and typical market conditions.

Tail risk spillovers among Chinese stock market sectors

This study employs a combination of the CAViaR model and the TVP-VAR-based connectedness approach to investigate tail risk spillovers among Chinese stock sectors. Using daily data on ten sector indices from January 5, 2006, to September 28, 2023, the empirical findings reveal significant time-varying tail risk spillovers among sectors, with heightened spillovers observed during extreme events. Moreover, the industrials, materials, and consumer discretionary sectors are found to be the senders of tail risk spillovers, while the energy, estate, and finance sectors are receivers during the sample period. The findings carry substantial implications for policy shaping and investment decision-making.

Chance-constrained UAM traffic flow optimization with fast disruption recovery under uncertain waypoint occupancy time

Trajectory-based operations offer a promising solution for effective urban air mobility (UAM) traffic management with conflict-free four-dimensional trajectories. However, these trajectories generated in strategic phases by existing methods could be significantly disrupted due to uncertainties such as flight delays and trajectory deviations. This paper develops a chance-constrained UAM traffic flow management (UTFM) optimization model with fast disruption recovery to solve observed disrupted trajectories and improve their resilience. Our model introduces a novel concept of waypoint occupancy time to cope with the probabilistic separation constraints induced by variables, notably urban wind field. We then convert the probabilistic constraint into a deterministic one, incorporating risk-bounded separation guarantees derived from flight experiment data. Furthermore, we develop a hierarchical stochastic search algorithm for solving the redefined deterministic optimization problem. Our comprehensive numerical studies demonstrate the model's effectiveness in restoring disrupted flights and resolving conflicts within seconds. Additionally, our reliability testing showcases the resilience of the model in managing disruptions, even at levels as high as 50%, and its adaptability in addressing varying levels of uncertainty risk. We further demonstrate the ability of the UTFM model to capture tail risks across Gaussian and non-Gaussian uncertainty distributions. Lastly, our scalability analysis highlights the potential capacity of the model to support up to 3,000 flights per hour in Singapore's urban airspace below 400 ft. This study introduces an adaptable framework to facilitate the modelling of robust traffic flow management and performance-based separation for a variety of eVTOL types under uncertainties.

On two new modified tawn copulas

At its core, copula theory focuses on constructing a copula function, which characterizes the structure of dependence between random variables. In particular, the creation of extreme value copulas is crucial because they allow accurate modeling of extreme dependence that traditional copulas can ignore. In this article, we propose theoretical developments on this subject by proposing two new extreme value copulas. They aim to extend the functionalities of the so-called Tawn copula. This is of interest because the Tawn copula is known to be a powerful tool in modeling joint distributions, particularly in capturing asymmetric and upper tail dependences, making it valuable for analyzing extreme events and tail risk. The proposed copulas are designed to go beyond these attractive features. On the mathematical side, they are derived from new Pickands dependence functions; one modifies the Pickands dependence function of the Tawn copula by using a polynomial-exponential function, and the other does the same but by introducing a power function. The proofs are based on diverse differentiation, arrangement, and inequality techniques. Overall, the created copulas are attractive because (i) they possess modulable levels of asymmetry, (ii) they depend on several tuning parameters, making them very flexible in terms of upper tail dependence in particular, and (iii) they benefit from interesting correlation ranges of values. Several figures and value tables support the theoretical findings.

Nonlinear behavior of tail risk resonance and early warning: Insight from global energy stock markets

This article calculates the tail risk of the global energy market and explores the nonlinear behavior of tail risk resonance at the global, regional and national levels from a dynamic perspective. Moreover, we apply various machine learning models to perform the experimental training on the drivers that affect tail risk resonance to establish a risk warning system. By simulation training and evaluation, we screen the optimal risk warning model and identify the important drivers of risk resonance in the energy market. The findings indicate that the energy system has a strong tail risk resonance phenomenon, and the occurrence of extreme events as well as a collapse in oil prices can significantly increase the system's risk resonance. The European region is the dominant risk emitter and the American region is the main risk receiver. The Russian-Ukrainian war significantly increased the risk resonance across regions. The Argentine energy market is the main receiver of tail risk and the US energy market is the primary sender of tail risk. By incorporating the risk factors into several machine learning models, the Random Forest model is found to provide effective early warning of tail risks. Additionally, importance analysis reveals that inflation levels, stock market volatility and trade policy changes are important factors affecting risk resonance in the energy market and have nonlinear effects on the risk resonance of the energy system.

Are “too big to fail” banks just different in size? – A study on systemic risk and stand-alone risk

This study shows that investment decisions drive tail risks (i.e., systemic risk and stand-alone tail risk) of TBTF (Too-Big-to-Fail) banks, while financing decisions determine tail risks of non-TBTF banks. After the Dodd-Frank Act, undercapitalized non-TBTF banks continue to gamble for resurrection, and their stand-alone tail risk become more sensitive to funding availability and net-stable-funding-ratio than TBTF banks. We show that implementing a slimmed-down version of TBTF regulations on non-TBTF banks cannot efficiently contain the stand-alone risk of non-TBTF banks and cannot eliminate TBTF privilege. Moreover, non-TBTF banks together generate larger pressure of contagion on the real economy, and they herd more when making financing decisions after the Act. Our findings highlight the need for enhanced regulations on the liability-side of non-TBTF banks.

Is macroeconomic tail risk contagious to stock idiosyncratic risk?

We explore whether macroeconomic tail risk (MTR) is contagious at the individual stock level. We find that macroeconomic tail risk exacerbates stock idiosyncratic volatility. An increase in macroeconomic tail risk by one standard deviation is associated with an average increase in idiosyncratic volatility by 5.10%. Further, divergence of opinion intensifies the macro to stock idiosyncratic risk contagion. Results show that the positive impact of macroeconomic tail risk on idiosyncratic volatility is only observed in assets with positive MTR beta, indicating that risk-averse hedging trading behavior of investors during macroeconomic downturns is the potential reason for the macro-individual risk contagion.

Bitcoin replication using machine learning

Cryptocurrencies are characterized by high volatility and low correlations with traditional asset classes, and present an intriguing investment opportunity. However, their inherent risks and regulatory uncertainties make direct investment challenging for many investors. This paper addresses this challenge by proposing a replication framework that employs machine learning to create synthetic portfolios that replicate the risk-adjusted return profile and diversification benefits of Bitcoin, by far the largest cryptocurrency by market share. We show that the synthetic portfolios offer a compelling alternative to direct investment in Bitcoin, delivering superior risk-adjusted returns net of trading costs while mitigating the risks that are associated with holding Bitcoin directly. Furthermore, the synthetic portfolios provide better diversification benefits and lower tail risk.

Forecasting the realized volatility of agricultural commodity prices: Does sentiment matter?

AbstractWe analyze the out‐of‐sample predictive power of sentiment for the realized volatility of agricultural commodity price returns. We use high‐frequency intra‐day data covering the period from 2009 to 2020 to estimate realized volatility. Our baseline forecasting model is a heterogeneous autoregressive (HAR) model, which we extend to include sentiment. We further enhance this model by incorporating various key realized moments such as leverage, realized skewness, realized kurtosis, realized upside (“good”) volatility, realized downside (“bad”) volatility, realized jumps, realized upside tail risk, and realized downside tail risk. In order to setup a forecasting model, we use (i) forward and backward stepwise predictor selection and (ii) a model‐based averaging algorithm. The forecasting models constructed through these algorithms outperform both the baseline HAR‐RV model and the HAR‐RV‐sentiment model. We conclude that, for the agricultural commodities studied in our research, realized moments play a more significant role in forecasting realized volatility compared to sentiment.