|
on Risk Management |
|
Issue of 2022‒06‒27
24 papers chosen by |
| By: | Daniel Dimitrov (University of Amsterdam); Sweder van Wijnbergen (University of Amsterdam) |
| Abstract: | We propose a credit portfolio approach for evaluating systemic risk and attributing it across institutions. We construct a model that can be estimated from high-frequency CDS data. This captures risks from privately held institutions and cooperative banks, extending approaches that rely on information from the public equity market. We account for correlated losses between the institutions, overcoming a modeling weakness in earlier studies. A latent risk factor with heterogeneous exposures fitted on the implied default probabilities quantifies the potential for joint distress and losses. We apply the model to a universe of Dutch banks and insurers. |
| Keywords: | Systemic risk, CDS rates, implied market measures, financial institutions |
| JEL: | G01 G20 G18 G38 |
| Date: | 2022–05–28 |
| URL: | http://d.repec.org/n?u=RePEc:tin:wpaper:20220034&r= |
| By: | Ivan De Lorenzo Buratta; Marina Feliciano; Duarte Maia |
| Abstract: | We quantify the effect of cyclical systemic risk and economic sentiment on non-financial corporations and households’ (total) credit growth for Portugal between 1991Q1 and 2020Q2, following the Growth-at-risk methodology. We focus on the right-hand tail of the future credit growth distribution, as credit booms are potentially detrimental to financial stability. A set of measures of the upside tail risk in credit growth is computed to provide policymakers with more information to anticipate credit build-ups. We find that financial vulnerabilities and industrial sector economic confidence increase the upper tail risk of credit growth realizations for non-financial corporations in the short term (4 quarters horizon). At the medium to long term (12 quarters horizon), the impact of those indicators almost cancels each other out. As regards households, increasing financial vulnerabilities and consumers’ economic confidence display opposite effects on the upper tail risk of credit growth, at short and medium to long terms. Credit-at-risk anticipates credit build-ups preceding financial crises and decelerations corresponding to recessions. The upper tail to median and the upper to lower tail distances identify the upper tail dynamics as the main responsible for future credit growth uncertainty. Expected longrise reinforces Credit-at-risk results while the probabilities of observing future credit growth above its mean and credit growth one standard deviation above its current value exhibit high levels before 2008 for both non-financial corporations and households, followed by deep falls during recessions which signal credit busts. For all the measures, the 2013-2018 increase in tail risk depends on the structural change in credit growth dynamics observed in the early 2000s. The most recent results highlight the predominant role of confidence indicators, further dampened in 2020 by the COVID-19 effects on the economic outlook. |
| JEL: | A1 |
| Date: | 2022 |
| URL: | http://d.repec.org/n?u=RePEc:ptu:wpaper:w202207&r= |
| By: | Ivan De Lorenzo Buratta; Marina Feliciano; Duarte Maia |
| Abstract: | By monitoring the evolution of risks to economic activity over time, we quantify the likelihood and severity of future negative economic growth. Following the Growth-at-risk approach, we explore the non-linear relationship between the current financial situation and the distribution of future GDP growth for Portugal. We find that both financial vulnerability and risk have a negative effect on the left tail of the one-year-ahead GDP growth distribution. Financial vulnerability has the largest impact on GDP growth at the medium to long term horizon while financial risk is only significant at the short term horizon. The GDP-at-risk measure signals economic recessions, no matter whether fueled by financial stress or imbalances, reaching negative values before 2008 and stagnating at low levels before the European Sovereign Debt Crisis. To provide policymakers with better tools to signal an increase in the likelihood of a crisis, we compute a set of complementary risk measures. Among those analyzed, the distance between the tails of the conditional distribution of GDP growth complements GDP-at-risk in anticipating economic recessions since it signals the Great Financial Crisis with a clear downward trend before 2008. The moments of the GDP growth distribution have some power in signalling recessions, as they identify changes in the characteristics of the distribution. Finally, we argue that the expected shortfall and longrise can complement the GDP-at-risk measure since they encompass information which is not limited to a single percentile of the distribution. |
| JEL: | C53 E01 E17 E27 E32 E44 G01 |
| Date: | 2022 |
| URL: | http://d.repec.org/n?u=RePEc:ptu:wpaper:w202204&r= |
| By: | Wing Fung Chong; Runhuan Feng; Hins Hu; Linfeng Zhang |
| Abstract: | Cyber risk is an omnipresent risk in the increasingly digitized world that is known to be difficult to quantify and assess. Despite the fact that cyber risk shows distinct characteristics from conventional risks, most existing models for cyber risk in the insurance literature have been purely based on frequency-severity analysis, which was developed for classical property and casualty risks. In contrast, the cybersecurity engineering literature employs different approaches, under which cyber incidents are viewed as threats or hacker attacks acting on a particular set of vulnerabilities. There appears a gap in cyber risk modeling between engineering and insurance literature. This paper presents a novel model to capture these unique dynamics of cyber risk known from engineering and to model loss distributions based on industry loss data and a particular company's cybersecurity profile. The analysis leads to a new tool for allocating resources of the company between cybersecurity investments and loss-absorbing reserves. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.08435&r= |
| By: | Eduardo Ramos-P\'erez; Pablo J. Alonso-Gonz\'alez; Jos\'e Javier N\'u\~nez-Vel\'azquez |
| Abstract: | In general insurance companies, a correct estimation of liabilities plays a key role due to its impact on management and investing decisions. Since the Financial Crisis of 2007-2008 and the strengthening of regulation, the focus is not only on the total reserve but also on its variability, which is an indicator of the risk assumed by the company. Thus, measures that relate profitability with risk are crucial in order to understand the financial position of insurance firms. Taking advantage of the increasing computational power, this paper introduces a stochastic reserving model whose aim is to improve the performance of the traditional Mack's reserving model by applying an ensemble of Recurrent Neural Networks. The results demonstrate that blending traditional reserving models with deep and machine learning techniques leads to a more accurate assessment of general insurance liabilities. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.07334&r= |
| By: | Mohammadreza Mahmoudi |
| Abstract: | This paper aims to analyze the effect of Bitcoin on portfolio optimization using mean-variance, conditional value-at-risk (CVaR), and Markov regime switching approaches. I assessed each approach and developed the next based on the prior approach's weaknesses until I ended with a high level of confidence in the final approach. Though the results of mean-variance and CVaR frameworks indicate that Bitcoin improves the diversification of a well-diversified international portfolio, they assume that assets' returns are developed linearly and normally distributed. However, the Bitcoin return does not have both of these characteristics. Due to this, I developed a Markov regime switching approach to analyze the effect of Bitcoin on an international portfolio performance. The results show that there are two regimes based on the assets' returns: 1- bear state, where returns have low means and high volatility, 2- bull state, where returns have high means and low volatility. |
| Date: | 2022–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.00335&r= |
| By: | Claudio Albanese (LaMME, ENSIIE); Cyril B\'en\'ezet (LaMME, ENSIIE); St\'ephane Cr\'epey (LPSM) |
| Abstract: | We revisit Burnett (2021b,a)'s notion of hedging valuation adjustment (HVA) in the direction of model risk. The resulting HVA can be seen as the bridge between a global fair valuation model and the local models used by the different desks of the bank. However, model risk and dynamic hedging frictions, such as transaction costs {\`a} la Burnett (2021b,a), indeed deserve a reserve, but a risk-adjusted one, so not only an HVA, but also a contribution to the KVA of the bank. We also argue that the industry-standard XVA metrics are jeopardized by cash flows risk, which is in fact of the same mathematical nature than the one regarding pricing models, although at the higher level of aggregation characteristic of XVA metrics. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.11834&r= |
| By: | Raifu, Isiaka Akande; Ogbonna, Ahamuefula E |
| Abstract: | The study assessed the hedge or safe-haven property of five cryptocurrencies for stocks of three COVID-19 worst-hit African countries. We address two main concerns bordering on the predictive capacity of African stocks for cryptocurrency returns, and the safe-haven property that cryptocurrencies could offer to African stocks. A distributed lag model, with explicitly incorporated salient statistical features, was adopted based its efficient management of parameter proliferation and estimation biases. We ascertained the model’s in-sample predictability and evaluate its out-of-sample forecasts performance in comparison with historical average model, using Clark and West statistics. While African stocks significantly predicted cryptocurrency returns, the cryptocurrency-stocks nexus revealed the diversifier and safe-haven property of cryptocurrencies for African stocks in periods of normalcy and crisis/pandemic, respectively. Our predictive model outperformed the historical average model in the out-of-sample. Our results may be sensitive to cryptocurrency-stocks nexus, sample periods but not the out-of-sample forecast horizons. |
| Keywords: | COVID-19; Cryptocurrency; Distributed Lag Model; Hedge; Safe-Haven |
| JEL: | C51 C58 G11 |
| Date: | 2021–01–19 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:113139&r= |
| By: | Peter Friz; Jim Gatheral |
| Abstract: | In this non-technical introduction to diamond trees and forests, we focus on their application to computation in stochastic volatility models written in forward variance form, rough volatility models in particular. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.03741&r= |
| By: | Marc Schmitt |
| Abstract: | Artificial intelligence (AI) and machine learning (ML) have become vital to remain competitive for financial services companies around the globe. The two models currently competing for the pole position in credit risk management are deep learning (DL) and gradient boosting machines (GBM). This paper benchmarked those two algorithms in the context of credit scoring using three distinct datasets with different features to account for the reality that model choice/power is often dependent on the underlying characteristics of the dataset. The experiment has shown that GBM tends to be more powerful than DL and has also the advantage of speed due to lower computational requirements. This makes GBM the winner and choice for credit scoring. However, it was also shown that the outperformance of GBM is not always guaranteed and ultimately the concrete problem scenario or dataset will determine the final model choice. Overall, based on this study both algorithms can be considered state-of-the-art for binary classification tasks on structured datasets, while GBM should be the go-to solution for most problem scenarios due to easier use, significantly faster training time, and superior accuracy. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.10535&r= |
| By: | Claudiu Vinte; Marcel Ausloos |
| Abstract: | To take into account the temporal dimension of uncertainty in stock markets, this paper introduces a cross-sectional estimation of stock market volatility based on the intrinsic entropy model. The proposed cross-sectional intrinsic entropy (CSIE) is defined and computed as a daily volatility estimate for the entire market, grounded on the daily traded prices: open, high, low, and close prices (OHLC), along with the daily traded volume for all symbols listed on The New York Stock Exchange (NYSE) and The National Association of Securities Dealers Automated Quotations (NASDAQ). We perform a comparative analysis between the time series obtained from the CSIE and the historical volatility as provided by the estimators: close-to-close, Parkinson, Garman-Klass, Rogers-Satchell, Yang-Zhang, and intrinsic entropy (IE), defined and computed from historical OHLC daily prices of the Standard & Poor's 500 index (S&P500), Dow Jones Industrial Average (DJIA), and the NASDAQ Composite index, respectively, for various time intervals. Our study uses approximately 6000 day reference points, starting on 1 Jan. 2001, until 23 Jan. 2022, for both the NYSE and the NASDAQ. We found that the CSIE market volatility estimator is consistently at least 10 times more sensitive to market changes, compared to the volatility estimate captured through the market indices. Furthermore, beta values confirm a consistently lower volatility risk for market indices overall, between 50% and 90% lower, compared to the volatility risk of the entire market in various time intervals and rolling windows. |
| Date: | 2022–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.00104&r= |
| By: | Victor Olkhov |
| Abstract: | This paper introduces the market-based asset price probability during time averaging interval $\Delta$. We substitute the present problem of guessing the "correct" form of the asset price probability by description of the price probability as function of the market trade value and volume statistical moments during $\Delta$. We define n-th price statistical moments as ratio of n-th statistical moments of the trade value to n-th statistical moments of the trade volume. That definition states no correlations between time-series of n-th power of the trade volume and price during $\Delta$, but doesn't result statistical independence between the trade volume and price. The set of price n-th statistical moments defines Taylor series of the price characteristic function. Approximations of the price characteristic function that reproduce only first m price statistical moments, generate approximations of the market-based price probability. That approach unifies probability description of market-based asset price, price indices, returns, inflation and their volatilities. Market-based price probability approach impacts the asset pricing models and uncovers hidden troubles and usage bounds of the widespread risk hedging tool -- Value-at-Risk, lets you determine the price autocorrelations and revises the classical option pricing from one to two dimensional problem. Market-based approach doesn't simplify the price probability puzzle but establishes direct economic ties between asset pricing, market randomness and economic theory. Description of the market-based price and returns volatility, Skewness and Kurtosis requires development of economic theories those model relations between second, third and forth order macroeconomic variables. Development of these theories will take a lot of efforts and years. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.07256&r= |
| By: | International Association of Deposit Insurers |
| Keywords: | deposit insurance, bank resolution |
| JEL: | G21 G33 |
| Date: | 2021–07 |
| URL: | http://d.repec.org/n?u=RePEc:awl:stands:2107&r= |
| By: | Simon M. S. Lo; Ralf A. Wilke |
| Abstract: | A typical situation in competing risks analysis is that the researcher is only interested in a subset of risks. This paper considers a depending competing risks model with the distribution of one risk being a parametric or semi-parametric model, while the model for the other risks being unknown. Identifiability is shown for popular classes of parametric models and the semiparametric proportional hazards model. The identifiability of the parametric models does not require a covariate, while the semiparametric model requires at least one. Estimation approaches are suggested which are shown to be $\sqrt{n}$-consistent. Applicability and attractive finite sample performance are demonstrated with the help of simulations and data examples. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.06087&r= |
| By: | A. Max Reppen; H. Mete Soner; Valentin Tissot-Daguette |
| Abstract: | A method based on deep artificial neural networks and empirical risk minimization is developed to calculate the boundary separating the stopping and continuation regions in optimal stopping. The algorithm parameterizes the stopping boundary as the graph of a function and introduces relaxed stopping rules based on fuzzy boundaries to facilitate efficient optimization. Several financial instruments, some in high dimensions, are analyzed through this method, demonstrating its effectiveness. The existence of the stopping boundary is also proved under natural structural assumptions. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.04595&r= |
| By: | Nakajima, Jouchi |
| Abstract: | A stock market volatility index is a widely-used proxy of uncertainty in the macroeconomy, and its increase is shown to dampen real economic activity. In contrast, the macroeconomic uncertainty index proposed by Jurado et al. (2015) measures the predictability of a wide range of macroeconomic indicators and thus is a comprehensive indicator of macroeconomy-wide uncertainty. This paper empirically investigates a nonlinear link between financial volatility and real economic activity depending on the level of the macroeconomic uncertainty index. Based on the United States and Japan data, empirical analysis suggests that an increase in the financial volatility lowers industrial production and business fixed investment more persistently when the macroeconomic uncertainty is higher. |
| Keywords: | Financial volatility, Macroeconomic uncertainty, Nonlinear effect |
| JEL: | E32 E52 |
| Date: | 2022–06 |
| URL: | http://d.repec.org/n?u=RePEc:hit:hiasdp:hias-e-121&r= |
| By: | Tian Chen; Ruyi Liu; Zhen Wu |
| Abstract: | In this paper, we consider a continuous-time mean-variance portfolio selection with regime-switching and random horizon. Unlike previous works, the dynamic of assets are described by non-Markovian regime-switching models in the sense that all the market parameters are predictable with respect to the filtration generated jointly by Markov chain and Brownian motion. We formulate this problem as a constrained stochastic linear-quadratic optimal control problem. The Markov chain is assumed to be independent of the Brownian motion. So the market is incomplete. We derive closed-form expressions for both the optimal portfolios and the efficient frontier. All the results are different from those in the problem with fixed time horizon. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.06434&r= |
| By: | Keiko Fukumori; Ayumi Arai; Tomoya Matsumoto |
| Abstract: | This study examines the determinants of smallholder farmers’ adoption of weather-index crop insurance, which is considered to be a promising means of mitigating the negative welfare impacts of crop loss caused by drought or excess rainfall. The study utilizes household survey data covering 495 smallholder farmers in rural Kenya. It finds that a better understanding of insurance, together with a significant positive effect of years of education, considerably increases insurance uptake. The evidence suggests that it is important to provide educational programs on new financial products when introducing such products to smallholder farmers. However, it also shows the limitations of this study by revealing how important proper study design is to draw reliable methodological impact evaluations. |
| Keywords: | agriculture, weather risk, weather-index insurance, rural households, Kenya, JEL (O12, O13, O33, G22) |
| Date: | 2022–04 |
| URL: | http://d.repec.org/n?u=RePEc:jic:wpaper:230&r= |
| By: | Christopher J. Waller |
| Date: | 2022–06–03 |
| URL: | http://d.repec.org/n?u=RePEc:fip:fedgsq:94334&r= |
| By: | Kick, Andreas; Rottmann, Horst |
| Abstract: | Banks have always played an ambivalent role in financial markets. On the one hand, they provide essential services for the market; on the other hand, problems in the banking sector can send shock waves through the entire economy. Given this prominent role, it is not surprising that Pereira and Rua (2018) found that the health of the banking sector exerts an influence on stock returns in the US. Understanding the relationship between banks and their impact on the asset prices of non-financials is essential to evaluate the risk emanating from an unhealthy banking sector and should be considered in new regulatory requirements. The aim of this study is to determine if the health of European banks is of such importance for the European stock market so that spillover effects are visible. Our results show that none of our banking-health variables have explanatory power on the cross-section of European stock returns. These findings contrast those for the US. The reasons may be manifold, from an unimportant liquidity provisioning channel over reduced room for actions due to regulatory requirements up to a moral hazard situation in Europe, where investors strongly rely on the governmental bailouts of distressed banks. |
| Keywords: | asset pricing,banking,spillover,errors-in-variables,individual stocks,distance-to-default |
| JEL: | G12 G21 |
| Date: | 2022 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:hawdps:84&r= |
| By: | Christian Cox; Akanksha Negi; Digvijay Negi |
| Abstract: | In a world with costly transfers, some agents with high transaction costs may not find it feasible to trade. Hence, their consumption will co-vary with their endowment, leading to imperfect risksharing. In this paper, we augment the canonical risk-sharing model to incorporate frictions in the form of transaction costs. In this augmented model, given a particular network structure, risk sharing will happen within networks and not the whole universe of agents. We show that transaction costs and the implied network structure of trade have important implications for the tests of risk sharing. Using this model, we derive a structural risk-sharing test that uses consumption and production data alongside the trade network structure. We implement our method using data from the global trade of three major staple food commodities. Comparing our estimates with the benchmark of frictionless trade, we find some evidence of transaction costs impeding risk-sharing in these commodities. |
| Date: | 2022 |
| URL: | http://d.repec.org/n?u=RePEc:msh:ebswps:2022-5&r= |
| By: | Minzarari, Dumitru |
| Abstract: | The ongoing military and political escalations in and around Donbas - including the increase in Russian military deployments near Ukraine's borders - represent one of the most severe security crises in Europe since Russia's aggression against Ukraine in 2014. The patterns of Russian military deployments, the structure of forces, and the types of observed military hardware strongly suggest the risk of an offensive operation rather than an exercise. Given the existing political costs, that operation is likely to take indirect forms by using the cover of Russian military proxies in Donbas. This crisis represents both a major challenge and an opportunity for the European Union (EU) to conduct practical work on developing its strategic autonomy and offer leadership in strengthening the security in its immediate neighbourhood. What should the EU do in practical terms to discourage further military escalation around Donbas, or at least increase the costs for such a development? |
| Date: | 2021 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:swpcom:272021&r= |
| By: | Andrew E. Clark; Conchita D'Ambrosio; Anthony Lepinteur |
| Abstract: | Job insecurity is one of the risks that workers face on the labour market. As with any risk, individuals can choose to insure against it. We here consider marriage as a way of insuring against labour-market risk. The 1999 rise in the French Delalande tax, paid by large private firms when they laid off workers aged 50 or over, led to an exogenous rise in job insecurity for the uncovered (younger workers) in the affected firms. A difference-in-differences analysis using French panel data reveals that this greater job insecurity for the under-50s led to a significant rise in their probability of marriage, and especially when the partner had greater job security, consistent with marriage providing insurance against labour-market risk. |
| Keywords: | marriage, insurance, employment protection, perceived job security, difference-in-differences |
| Date: | 2021–06–30 |
| URL: | http://d.repec.org/n?u=RePEc:cep:cepdps:dp1778&r= |
| By: | Rafael Reisenhofer; Xandro Bayer; Nikolaus Hautsch |
| Abstract: | Despite the impressive success of deep neural networks in many application areas, neural network models have so far not been widely adopted in the context of volatility forecasting. In this work, we aim to bridge the conceptual gap between established time series approaches, such as the Heterogeneous Autoregressive (HAR) model, and state-of-the-art deep neural network models. The newly introduced HARNet is based on a hierarchy of dilated convolutional layers, which facilitates an exponential growth of the receptive field of the model in the number of model parameters. HARNets allow for an explicit initialization scheme such that before optimization, a HARNet yields identical predictions as the respective baseline HAR model. Particularly when considering the QLIKE error as a loss function, we find that this approach significantly stabilizes the optimization of HARNets. We evaluate the performance of HARNets with respect to three different stock market indexes. Based on this evaluation, we formulate clear guidelines for the optimization of HARNets and show that HARNets can substantially improve upon the forecasting accuracy of their respective HAR baseline models. In a qualitative analysis of the filter weights learnt by a HARNet, we report clear patterns regarding the predictive power of past information. Among information from the previous week, yesterday and the day before, yesterday's volatility makes by far the most contribution to today's realized volatility forecast. Moroever, within the previous month, the importance of single weeks diminishes almost linearly when moving further into the past. |
| Date: | 2022–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2205.07719&r= |