|
on Risk Management |
|
Issue of 2024‒06‒24
twelve papers chosen by |
| By: | Yuji Sakurai; Zhuohui Chen |
| Abstract: | We propose a new machine-learning-based approach for forecasting Value-at-Risk (VaR) named CoFiE-NN where a neural network (NN) is combined with Cornish-Fisher expansions (CoFiE). CoFiE-NN can capture non-linear dynamics of high-order statistical moments thanks to the flexibility of a NN while maintaining interpretability of the outputs by using CoFiE which is a well-known statistical formula. First, we explain CoFiE-NN. Second, we compare the forecasting performance of CoFiE-NN with three conventional models using both Monte Carlo simulation and real data. To do so, we employ Long Short-Term Memory (LSTM) as our main specification of the NN. We then apply the CoFiE-NN for different asset classes, with a focus on foreign exchange markets. We report that CoFiE-NN outperfoms the conventional EGARCH-t model and the Extreme Value Theory model in several statistical criteria for both the simulated data and the real data. Finally, we introduce a new empirical proxy for tail risk named tail risk ratio under CoFiE-NN. We discover that the only 20 percent of tail risk dynamics across 22 currencies is explained by one common factor. This is contrasting to the fact that 60 percent of volatility dynamics across the same currencies is explained by one common factor. |
| Keywords: | Machine learning; Value-at-Risk; Neural Network |
| Date: | 2024–05–10 |
| URL: | http://d.repec.org/n?u=RePEc:imf:imfwpa:2024/099&r= |
| By: | Martin Herdegen; Nazem Khan; Cosimo Munari |
| Abstract: | Risk and utility functionals are fundamental building blocks in economics and finance. In this paper we investigate under which conditions a risk or utility functional is sensitive to the accumulation of losses in the sense that any sufficiently large multiple of a position that exposes an agent to future losses has positive risk or negative utility. We call this property sensitivity to large losses and provide necessary and sufficient conditions thereof that are easy to check for a very large class of risk and utility functionals. In particular, our results do not rely on convexity and can therefore also be applied to most examples discussed in the recent literature, including (non-convex) star-shaped risk measures or S-shaped utility functions encountered in prospect theory. As expected, Value at Risk generally fails to be sensitive to large losses. More surprisingly, this is also true of Expected Shortfall. By contrast, expected utility functionals as well as (optimized) certainty equivalents are proved to be sensitive to large losses for many standard choices of concave and nonconcave utility functions, including $S$-shaped utility functions. We also show that Value at Risk and Expected Shortfall become sensitive to large losses if they are either properly adjusted or if the property is suitably localized. |
| Date: | 2024–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2405.12154&r= |
| By: | Yu Cheng; Qin Yang; Liyang Wang; Ao Xiang; Jingyu Zhang |
| Abstract: | In the realm of globalized financial markets, commercial banks are confronted with an escalating magnitude of credit risk, thereby imposing heightened requisites upon the security of bank assets and financial stability. This study harnesses advanced neural network techniques, notably the Backpropagation (BP) neural network, to pioneer a novel model for preempting credit risk in commercial banks. The discourse initially scrutinizes conventional financial risk preemptive models, such as ARMA, ARCH, and Logistic regression models, critically analyzing their real-world applications. Subsequently, the exposition elaborates on the construction process of the BP neural network model, encompassing network architecture design, activation function selection, parameter initialization, and objective function construction. Through comparative analysis, the superiority of neural network models in preempting credit risk in commercial banks is elucidated. The experimental segment selects specific bank data, validating the model's predictive accuracy and practicality. Research findings evince that this model efficaciously enhances the foresight and precision of credit risk management. |
| Date: | 2024–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2405.10762&r= |
| By: | Financial System and Bank Examination Department (Bank of Japan); Strategy Development and Management Bureau (Financial Services Agency) |
| Abstract: | Securing stable foreign currency liquidity is one of the most important issues for Japanese major banks, as it is the basis of the expansion of their overseas businesses. The March 2023 banking turmoil in the United States and Switzerland shed new light on the importance of managing liquidity risk. Against this background, major banks have been enhancing their risk management through foreign currency liquidity stress testing based on more conservative and appropriate stress scenarios, early warning frameworks, and prompt and accurate liquidity data management. The Financial Services Agency and the Bank of Japan have supported these efforts through initiatives including joint surveys. As a result, major banks' resilience to foreign currency liquidity risk has steadily improved. However, there remains room for further enhancement. Going forward, banks are expected to continue their efforts to further enhance their risk management in line with changes in the risk profiles of their overseas businesses and the external environment. |
| Date: | 2024–05–22 |
| URL: | http://d.repec.org/n?u=RePEc:boj:bojrev:rev24e03&r= |
| By: | Ngoepe, Letlhogonolo Kearabilwe; Bonga-Bonga, Lumengo |
| Abstract: | Various ‘green’ investment channels cater specifically to environmentally conscious investments. In this paper, we investigate the optimal green investment strategy by comparing the risk of three green financial instruments– green bonds, green equity, and a balanced 50/50 bond equity fund. Using the dynamic and frequency connectedness approaches by Diebold and Yilmaz (2012) and Baruník and Křehlík (2018), we analyze how financial risk affects green investment over various time horizons. Our findings show that green equity possesses the highest risk spillovers. Furthermore, green bonds and the ESG equity index provide risk diversification benefits for green investors. The balanced index displays a low risk-return nexus, further indicating that green investors are better off by investing in a diversified portfolio. Lastly, under unfavourable market conditions, the green investment market instruments provide little to no diversification against each other. |
| Keywords: | Green equity, ESG equity index, balanced index, frequency connetedness |
| JEL: | C5 F3 G15 |
| Date: | 2024 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:121091&r= |
| By: | Xuewei Zhou (School of Finance, Shanghai University of Finance and Economics, and Shanghai Institute of International Finance and Economics, Shanghai 200433, China); Zisheng Ouyang (Business School, Hunan Normal University, and Hunan Key Laboratory of Macroeconomic Big Data Mining and its Application, Changsha, Hunan 410081, China); Rangan Gupta (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Qiang Ji (Institutes of Science and Development, Chinese Academy of Sciences, Beijing 100190, China) |
| Abstract: | This paper constructs multilayer frequency networks containing short-, medium-, and long-term layers to examine the frequency connectedness among commodity futures markets. We examine the frequency heterogeneity of commodity volatility connectedness at the average, dynamic, and crisis levels. We also investigate the determinants of frequency connectedness among commodity futures markets. The results show that there are strong short-term volatility spillovers between commodity futures markets, while connectedness during crises is dominated by long-term factors. We find that there is heterogeneity in the edge structure of short- and long-term networks during the crisis. In addition, we note that cocoa futures can hedge frequency risk in commodity markets. Determinants analysis suggests that inflation risk is the key driver of frequency connectedness in commodity futures. Moreover, the drivers of connectedness differ between short-, medium-, and long-term. Our work provides new insights for studying the risk contagion of commodity markets and informs the decisions of investors and regulators. |
| Keywords: | Frequency connectedness, Multilayer networks, Commodity futures markets, Systemic risk |
| JEL: | G14 G15 G32 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:pre:wpaper:202422&r= |
| By: | Jiahao Weng; Yan Xie |
| Abstract: | In this study, we constructed daily high-frequency sentiment data and used the VAR method to attempt to predict the next day's implied volatility surface. We utilized 630, 000 text data entries from the East Money Stock Forum from 2014 to 2023 and employed deep learning methods such as BERT and LSTM to build daily market sentiment indicators. By applying FFT and EMD methods for sentiment decomposition, we found that high-frequency sentiment had a stronger correlation with at-the-money (ATM) options' implied volatility, while low-frequency sentiment was more strongly correlated with deep out-of-the-money (DOTM) options' implied volatility. Further analysis revealed that the shape of the implied volatility surface contains richer market sentiment information beyond just market panic. We demonstrated that incorporating this sentiment information can improve the accuracy of implied volatility surface predictions. |
| Date: | 2024–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2405.11730&r= |
| By: | Gang Hu; Ming Gu |
| Abstract: | Investment portfolios, central to finance, balance potential returns and risks. This paper introduces a hybrid approach combining Markowitz's portfolio theory with reinforcement learning, utilizing knowledge distillation for training agents. In particular, our proposed method, called KDD (Knowledge Distillation DDPG), consist of two training stages: supervised and reinforcement learning stages. The trained agents optimize portfolio assembly. A comparative analysis against standard financial models and AI frameworks, using metrics like returns, the Sharpe ratio, and nine evaluation indices, reveals our model's superiority. It notably achieves the highest yield and Sharpe ratio of 2.03, ensuring top profitability with the lowest risk in comparable return scenarios. |
| Date: | 2024–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2405.05449&r= |
| By: | Lutz Kilian; Michael D. Plante; Alexander W. Richter |
| Abstract: | This paper seeks to understand the general equilibrium effects of time-varying geopolitical risk in oil markets. Answering this question requires simultaneously modeling several features including macroeconomic disasters and geopolitically driven oil production disasters, oil storage and precautionary savings, and the endogenous determination of uncertainty about output and the price of oil. We find that oil price uncertainty tends to be driven by macroeconomic uncertainty. Shifts in the probability of a geopolitically driven major oil supply disruption have meaningful effects on the price of oil and the macro economy, but the resulting oil price uncertainty is not a major driver of fluctuations in macroeconomic aggregates. |
| Keywords: | geopolitical risk; macroeconomic risk; time-varying uncertainty; rare disasters; oil; endogeneity; shock propagation; economic fluctuations; precautionary savings; inventories |
| JEL: | E13 E22 E32 Q43 |
| Date: | 2024–05–28 |
| URL: | https://d.repec.org/n?u=RePEc:fip:feddwp:98321&r= |
| By: | Cullen S. Hendrix (Peterson Institute for International Economics) |
| Abstract: | This paper investigates whether the El Nino Southern Oscillation (ENSO)--the warming and cooling cycle in the central and eastern Pacific Ocean that affects both global atmospheric and ocean conditions--is a driver of geopolitical risk at the global scale. Using nonlinear cross-convergent mapping, a technique for characterizing causal relationships in dynamic systems, it finds ENSO is causally related to geopolitical risk at the global level, but that finding is not replicated at the country level for countries whose economies are most strongly influenced by ENSO cycles. Put differently, ENSO-related geopolitical risk is an emergent phenomenon evident only at the Earth system level. Then, using monthly observations of ENSO and geopolitical risk, the paper reports a curvilinear, contemporaneous relationship between ENSO and risk, with La Nina conditions associated with lessened geopolitical risk relative to El Nino and neutral climate conditions. The effects are statistically and substantively significant, and the relationship is demonstrated to be stronger in more recent decades (post-1990). The effect for geopolitical risk of transitioning from La Nina to neutral ENSO conditions is of similar magnitude to that of the outbreak of a major interstate war. |
| Keywords: | climate change, global warming, natural disasters, geopolitical risk, domestic and international conflicts |
| JEL: | C53 D74 F51 Q34 Q54 |
| Date: | 2024–05 |
| URL: | https://d.repec.org/n?u=RePEc:iie:wpaper:wp24-14&r= |
| By: | Marín Díazaraque, Juan Miguel; Romero, Eva; Lopes Moreira Da Veiga, María Helena |
| Abstract: | The paper proposes the use of Laplace approximation (LA) to estimate complex univariate symmetric and asymmetric stochastic volatility (SV) models with flexible distributions for standardized returns. LA is a method for approximating integrals, especially in Bayesian statistics, and is often used to approximate the posterior distribution of the model parameters. This method simplifies complex problems by focusing on the most critical areas and using a well-understood approximation. We show how easily complex SV models can be estimated and analyzed using LA, with changes to specifications, priors, and sampling error distributions requiring only minor changes to the code. The simulation study shows that the LA estimates of the model parameters are close to the true values in finite samples and that the proposed estimator is computationally efficient and fast. It is an effective alternative to existing estimation methods for SV models. Finally, we evaluate the in-sample and out-of-sample performance of the models by forecasting one-day-ahead volatility. We use four well-known energy index series: two for clean energy and two for conventional (brown) energy. In the out-of-sample analysis, we also examine the impact of climate policy uncertainty and energy prices on the volatility forecasts. The results support the use of asymmetric SV models for clean energy series and symmetric SV models for brown energy indices conditional on these state variables. |
| Keywords: | Asymmetric Volatility; Laplace Approximation; Stochastic Volatility |
| Date: | 2024–06–06 |
| URL: | https://d.repec.org/n?u=RePEc:cte:wsrepe:43947&r= |
| By: | Keyu Wu; Ernst Fehr; Sean Hofland; Martin Schonger |
| Abstract: | Ambiguous prospects are ubiquitous in social and economic life, but the psychological foundations of behavior under ambiguity are still not well understood. One of the most robust empirical regularities is the strong correlation between attitudes towards ambiguity and compound risk which suggests that compound risk aversion may provide a psychological foundation for ambiguity aversion. However, compound risk aversion and ambiguity aversion may also be independent psychological phenomena, but what would then explain their strong correlation? We tackle these questions by training a treatment group’s ability to reduce compound to simple risks, and analyzing how this affects their compound risk and ambiguity attitudes in comparison to a control group who is taught something unrelated to reducing compound risk. We find that aversion to compound risk disappears almost entirely in the treatment group, while the aversion towards both artificial and natural sources of ambiguity remain high and are basically unaffected by the teaching of how to reduce compound lotteries. Moreover, similar to previous studies, we observe a strong correlation between compound risk aversion and ambiguity aversion, but this correlation only exists in the control group while in the treatment group it is rather low and insignificant. These findings suggest that ambiguity attitudes are not a psychological relative, and derived from, attitudes towards compound risk, i.e., compound risk aversion and ambiguity aversion do not share the same psychological foundations. While compound risk aversion is primarily driven by a form of bounded rationality – the inability to reduce compound lotteries – ambiguity aversion is unrelated to this inability, suggesting that ambiguity aversion may be a genuine preference in its own right. |
| Date: | 2024–05 |
| URL: | https://d.repec.org/n?u=RePEc:zur:econwp:444&r= |