nep-rmg New Economics Papers
on Risk Management
Issue of 2023‒08‒28
thirteen papers chosen by
Stan Miles, Thompson Rivers University

  1. Reinforcement Learning for Credit Index Option Hedging By Francesco Mandelli; Marco Pinciroli; Michele Trapletti; Edoardo Vittori
  2. Labour at risk By Botelho, Vasco; Foroni, Claudia; Renzetti, Andrea
  3. Quantifying financial stability trade-offs for monetary policy: a quantile VAR approach By Chavleishvili, Sulkhan; Kremer, Manfred; Lund-Thomsen, Frederik
  4. Risk retention in the European securitization market: skimmed by the skin-in-the-game methods? By van Breemen, Vivian M.; Schwarz, Claudia; Vink, Dennis
  5. Adversarial Deep Hedging: Learning to Hedge without Price Process Modeling By Masanori Hirano; Kentaro Minami; Kentaro Imajo
  6. Beta-Sorted Portfolios By Matias D. Cattaneo; Richard K. Crump; Weining Wang
  7. An exploration of the mathematical structure and behavioural biases of financial crises By Nick James; Max Menzies
  8. The Anatomy of Cyber Risk By Rustam Jamilov; Helene Rey; Ahmed Tahoun
  9. Realized Stock Market Volatility of the United States: The Role of Employee Sentiment By Rangan Gupta; Savanah Hall; Christian Pierdzioch
  10. Good vs. Bad Volatility: The Dichotomy and Drivers of Connectedness in Major Cryptocurrencies By Jan Sila; Evzen Kocenda; Ladislav Kristoufek; Jiri Kukacka
  11. Including individual Customer Lifetime Value and competing risks in tree-based lapse management strategies By Mathias Valla; Xavier Milhaud; Anani Ayodélé Olympio
  12. VolTS: A Volatility-based Trading System to forecast Stock Markets Trend using Statistics and Machine Learning By Ivan Letteri
  13. Do You Even Crypto, Bro? Cryptocurrencies in Household Finance By Weber, Michael; Candia, Bernardo; Coibion, Olivier; Gorodnichenko, Yuriy

  1. By: Francesco Mandelli; Marco Pinciroli; Michele Trapletti; Edoardo Vittori
    Abstract: In this paper, we focus on finding the optimal hedging strategy of a credit index option using reinforcement learning. We take a practical approach, where the focus is on realism i.e. discrete time, transaction costs; even testing our policy on real market data. We apply a state of the art algorithm, the Trust Region Volatility Optimization (TRVO) algorithm and show that the derived hedging strategy outperforms the practitioner's Black & Scholes delta hedge.
    Date: 2023–07
  2. By: Botelho, Vasco; Foroni, Claudia; Renzetti, Andrea
    Abstract: We propose a Bayesian VAR model with stochastic volatility and time varying skewness to estimate the degree of labour at risk in the euro area and in the United States. We model the asymmetry of the shocks to changes in the unemployment rate as a function of real activity and financial risk factors. We find that the conditional distribution of the changes in the unemployment rate displays time-varying volatility and skewness, with peaks coinciding with the Global Financial Crisis and the COVID-19 pandemic. We take advantage of the multivariate nature of our parametric model to measure stagflation risk defined as the possible joint event of large increases in the unemployment rate and large annual rates of inflation. We find an increasing risk of stagflation for the euro area in 2022 while in the United States stagflation risk increased earlier in 2021 and started decreasing more recently. Notwithstanding the significantly high levels of inflation, stagflation risks have been contained by the resilient performance of the labour market in both areas. The degree of labour at risk is therefore important for the assessment of the inflation-unemployment trade-off. JEL Classification: C32, C53, E24, E27
    Keywords: Bayesian econometrics, labour market, stagflation risk, unemployment risk
    Date: 2023–08
  3. By: Chavleishvili, Sulkhan; Kremer, Manfred; Lund-Thomsen, Frederik
    Abstract: We propose a novel empirical approach to inform monetary policymakers about the potential effects of policy action when facing trade-offs between financial and macroeconomic stability. We estimate a quantile vector autoregression (QVAR) for the euro area covering the real economy, monetary policy and measures of ex ante and ex post systemic risk representing financial stability. Policy implications are derived from scenario analyses where the associated costs and benefits are functions of the projected paths of the potentially asymmetric distributions of inflation and economic growth, allowing us to take a risk management perspective. One exercise considers the intertemporal financial stability trade-off in the context of the global financial crisis, where we find ex post evidence in favour of monetary policy leaning against the financial cycle. Another exercise considers the short-term financial stability trade-off when deciding the appropriate speed of monetary policy tightening to combat inflationary pressures in a fragile financial environment. JEL Classification: C32, E37, E44, E52, G01
    Keywords: growth-at-risk, Policy trade-offs, quantile regression, systemic risk
    Date: 2023–07
  4. By: van Breemen, Vivian M.; Schwarz, Claudia; Vink, Dennis
    Abstract: We empirically investigated the impact of regulatory risk retention methods on credit ratings and pricing at issuance using a sample of European securitization tranches issued in the period 2011-2021. European regulation is based on the assumption that all risk retention methods homogenously align incentives and interests between originators and investors. We investigated the impact of these methods on the pricing of securitization tranches and found that investors adjust the risk premium at issuance for tranches based on different risk retention methods. We also found that credit ratings (discrepancy) differed depending on the risk retention method used. Finally, we gained a deeper insight into the risk retention methods chosen over time and concluded that originators take deal complexity and capital relief characteristics into consideration when selecting a specific method. JEL Classification: G12, G21, G24, G28
    Keywords: credit ratings, primary issuance spread, risk retention rule
    Date: 2023–08
  5. By: Masanori Hirano; Kentaro Minami; Kentaro Imajo
    Abstract: Deep hedging is a deep-learning-based framework for derivative hedging in incomplete markets. The advantage of deep hedging lies in its ability to handle various realistic market conditions, such as market frictions, which are challenging to address within the traditional mathematical finance framework. Since deep hedging relies on market simulation, the underlying asset price process model is crucial. However, existing literature on deep hedging often relies on traditional mathematical finance models, e.g., Brownian motion and stochastic volatility models, and discovering effective underlying asset models for deep hedging learning has been a challenge. In this study, we propose a new framework called adversarial deep hedging, inspired by adversarial learning. In this framework, a hedger and a generator, which respectively model the underlying asset process and the underlying asset process, are trained in an adversarial manner. The proposed method enables to learn a robust hedger without explicitly modeling the underlying asset process. Through numerical experiments, we demonstrate that our proposed method achieves competitive performance to models that assume explicit underlying asset processes across various real market data.
    Date: 2023–07
  6. By: Matias D. Cattaneo; Richard K. Crump; Weining Wang
    Abstract: Beta-sorted portfolios—portfolios comprised of assets with similar covariation to selected risk factors—are a popular tool in empirical finance to analyze models of (conditional) expected returns. Despite their widespread use, little is known of their statistical properties in contrast to comparable procedures such as two-pass regressions. We formally investigate the properties of beta-sorted portfolio returns by casting the procedure as a two-step nonparametric estimator with a nonparametric first step and a beta-adaptive portfolios construction. Our framework rationalizes the well-known estimation algorithm with precise economic and statistical assumptions on the general data generating process. We provide conditions that ensure consistency and asymptotic normality along with new uniform inference procedures allowing for uncertainty quantification and general hypothesis testing for financial applications. We show that the rate of convergence of the estimator is non-uniform and depends on the beta value of interest. We also show that the widely used Fama-MacBeth variance estimator is asymptotically valid but is conservative in general and can be very conservative in empirically relevant settings. We propose a new variance estimator, which is always consistent and provide an empirical implementation which produces valid inference. In our empirical application we introduce a novel risk factor—a measure of the business credit cycle—and show that it is strongly predictive of both the cross-section and time-series behavior of U.S. stock returns.
    Keywords: nonparametric estimation; partitioning; beta pricing models; portfolio sorting; partition; kernel regression; smoothly varying coefficients
    JEL: C12 C14 G12
    Date: 2023–07–01
  7. By: Nick James; Max Menzies
    Abstract: In this paper we contrast the dynamics of the 2022 Ukraine invasion financial crisis with notable financial crises of recent years - the dot-com bubble, global financial crisis and COVID-19. We study the similarity in market dynamics and associated implications for equity investors between various financial market crises and we introduce new mathematical techniques to do so. First, we study the strength of collective dynamics during different market crises, and compare suitable portfolio diversification strategies with respect to the unique number of sectors and stocks for optimal systematic risk reduction. Next, we introduce a new linear operator method to quantify distributional distance between equity returns during various crises. Our method allows us to fairly compare underlying stock and sector performance during different time periods, normalising for those collective dynamics driven by the overall market. Finally, we introduce a new combinatorial portfolio optimisation framework driven by random sampling to investigate whether particular equities and equity sectors are more effective in maximising investor risk-adjusted returns during market crises.
    Date: 2023–07
  8. By: Rustam Jamilov (All Souls College, University of Oxford); Helene Rey (London Business School); Ahmed Tahoun (London Business School)
    Abstract: This paper employs computational linguistics to introduce a novel text-based measure of firm-level cyber risk exposure based on quarterly earnings conference calls of listed firms. Our quarterly measures are available for more than 13, 000 firms from 85 countries over 2002-2021. We document that cyber risk exposure predicts cyber attacks, affects stock returns and profits, and is priced in the equity option market.The cost of option protection against price, variance, and tail risks is greater for more cyber-exposed firms. Cyber risks spill over across firms and persist at the sectoral level. The geography of cyber risk exposure is well approximated by a gravity model extended with cross-border portfolio flows. Back-of-the-envelope calculations suggest that the global cost of cyber risk is over $200 billion per year.
    Keywords: Cyber risk, textual analysis, earnings calls
    JEL: D22 L10
    Date: 2023–05–10
  9. By: Rangan Gupta (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Savanah Hall (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Christian Pierdzioch (Department of Economics, Helmut Schmidt University, Holstenhofweg 85, P.O.B. 700822, 22008 Hamburg, Germany)
    Abstract: Using data for the US stock market covering the sample period from 2008:06 to 2020:12, we study the incremental predictive value of employee sentiment for the realized volatility of stock returns. In doing so, we control for four different measures of investor sentiment and various macroeconomic and financial factors an uncertainties. We report results for several combination of forecast horizons and estimation windows and find that employee sentiment contributes to forecast accuracy for several of combinations.
    Keywords: Stock market volatility, Investor sentiment, Employee Sentiment, Forecasting
    JEL: C32 C53 G10
    Date: 2023–07
  10. By: Jan Sila (Charles University, Faculty of Social Sciences, Institute of Economic Studies, Prague, Czechia & Czech Academy of Sciences, Institute of Information Theory and Automation, Prague, Czechia); Evzen Kocenda (Charles University, Faculty of Social Sciences, Institute of Economic Studies, Prague, Czechia & Czech Academy of Sciences, Institute of Information Theory and Automation, Prague, Czechia); Ladislav Kristoufek (Charles University, Faculty of Social Sciences, Institute of Economic Studies, Prague, Czechia & Czech Academy of Sciences, Institute of Information Theory and Automation, Prague, Czechia); Jiri Kukacka (Charles University, Faculty of Social Sciences, Institute of Economic Studies, Prague, Czechia & Czech Academy of Sciences, Institute of Information Theory and Automation, Prague, Czechia)
    Abstract: Cryptocurrencies exhibit unique statistical and dynamic properties compared to those of traditional financial assets, making the study of their volatility crucial for portfolio managers and traders. We investigate the volatility connectedness dynamics of a representative set of eight major crypto assets. Methodologically, we decompose the measured volatility into positive and negative components and employ the time-varying parameters vector autoregression (TVP-VAR) framework to show distinct dynamics associated with market booms and downturns. The results suggest that crypto connectedness reflects important events and exhibits more variable and cyclical dynamics than those of traditional financial markets. Periods of extremely high or low connectedness are clearly linked to specific events in the crypto market and macroeconomic or monetary history. Furthermore, existing asymmetry from good and bad volatility indicates that information about market downturns spills over substantially faster than news about comparable market surges. Overall, the connectedness dynamics are predominantly driven by fundamental crypto factors, while the asymmetry measure also depends on macro factors such as the VIX index and the expected inflation.
    Keywords: Volatility, Dynamic connectedness, Asymmetric effects, Cryptocurrency
    JEL: C58 G10 C36
    Date: 2023–07
  11. By: Mathias Valla (LSAF - Laboratoire de Sciences Actuarielles et Financières [Lyon] - ISFA - Institut de Science Financière et d'Assurances, Faculty of Business and Economics - University of Leuven (KUL)); Xavier Milhaud (I2M - Institut de Mathématiques de Marseille - AMU - Aix Marseille Université - ECM - École Centrale de Marseille - CNRS - Centre National de la Recherche Scientifique); Anani Ayodélé Olympio (SAF - Laboratoire de Sciences Actuarielle et Financière - UCBL - Université Claude Bernard Lyon 1 - Université de Lyon)
    Abstract: A retention strategy based on an enlightened lapse model is a powerful profitability lever for a life insurer. Some machine learning models are excellent at predicting lapse, but from the insurer's perspective, predicting which policyholder is likely to lapse is not enough to design a retention strategy. In our paper, we define a lapse management framework with an appropriate validation metric based on Customer Lifetime Value and profitability. We include the risk of death in the study through competing risks considerations in parametric and tree-based models and show that further individualization of the existing approaches leads to increased performance. We show that survival tree-based models outperform parametric approaches and that the actuarial literature can significantly benefit from them. Then, we compare, on real data, how this framework leads to increased predicted gains for a life insurer and discuss the benefits of our model in terms of commercial and strategic decision-making.
    Keywords: Machine Learning, Life insurance, Customer lifetime value, Lapse, Lapse management strategy, Competing risks, Tree-based models
    Date: 2023
  12. By: Ivan Letteri
    Abstract: Volatility-based trading strategies have attracted a lot of attention in financial markets due to their ability to capture opportunities for profit from market dynamics. In this article, we propose a new volatility-based trading strategy that combines statistical analysis with machine learning techniques to forecast stock markets trend. The method consists of several steps including, data exploration, correlation and autocorrelation analysis, technical indicator use, application of hypothesis tests and statistical models, and use of variable selection algorithms. In particular, we use the k-means++ clustering algorithm to group the mean volatility of the nine largest stocks in the NYSE and NasdaqGS markets. The resulting clusters are the basis for identifying relationships between stocks based on their volatility behaviour. Next, we use the Granger Causality Test on the clustered dataset with mid-volatility to determine the predictive power of a stock over another stock. By identifying stocks with strong predictive relationships, we establish a trading strategy in which the stock acting as a reliable predictor becomes a trend indicator to determine the buy, sell, and hold of target stock trades. Through extensive backtesting and performance evaluation, we find the reliability and robustness of our volatility-based trading strategy. The results suggest that our approach effectively captures profitable trading opportunities by leveraging the predictive power of volatility clusters, and Granger causality relationships between stocks. The proposed strategy offers valuable insights and practical implications to investors and market participants who seek to improve their trading decisions and capitalize on market trends. It provides valuable insights and practical implications for market participants looking to.
    Date: 2023–07
  13. By: Weber, Michael (University of Chicago); Candia, Bernardo (University of California, Berkeley); Coibion, Olivier (University of Texas at Austin); Gorodnichenko, Yuriy (University of California, Berkeley)
    Abstract: Using repeated large-scale surveys of U.S. households, we study the cryptocurrency investment decisions and motives of households relative to other financial assets. Cryptocurrency holders tend to be young, white, male and more libertarian relative to non-crypto holders. They expect much higher rates of returns for crypto and perceive it as relatively safer than do other households. They also view it as a better hedge against inflation. For those holding cryptocurrencies, changes in Bitcoin prices translate into their purchases of durable goods. Finally, exogenously-provided information about historical returns of cryptocurrencies leads individuals to increase their desired crypto holdings and makes them more likely to actually purchase cryptocurrency subsequently. We compare these views and behaviors to those of households toward other financial assets and argue that cryptocurrency is unique in many of these respects.
    Keywords: cryptocurrency, household finance, surveys
    JEL: E4 G5 D8
    Date: 2023–07

This nep-rmg issue is ©2023 by Stan Miles. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.