
on Risk Management 
By:  Tobias Fissler; Jana Hlavinov\'a; Birgit Rudloff 
Abstract:  This paper is concerned with a twofold objective. Firstly, we establish elicitability and identifiability results for systemic risk measures introduced in Feinstein, Rudloff and Weber (2017). Specifying the entire set of capital allocations adequate to render a financial system acceptable, these systemic risk measures are examples of setvalued functionals. A functional is elicitable (identifiable) if it is the unique minimiser (zero) of an expected scoring function (identification function). Elicitability and identifiability are essential for forecast ranking and validation, $M$ and $Z$estimation, both possibly in a regression framework. To account for the setvalued nature of the systemic risk measures mentioned above, we secondly introduce a theoretical framework of elicitability and identifiability of setvalued functionals. It distinguishes between exhaustive forecasts, being setvalued and aiming at correctly specifying the entire functional, and selective forecasts, content with solely specifying a single point in the correct functional. Uncovering the structural relation between the two corresponding notions of elicitability and identifiability, we establish that a setvalued functional can be either selectively elicitable or exhaustively elicitable. Notably, selections of quantiles such as the lower quantile turn out not to be elicitable in general. Applying these structural results to systemic risk measures, we construct oriented selective identification functions, which induce a family of strictly consistent exhaustive elementary scoring functions. We discuss equivariance properties of these scores. We demonstrate their applicability in a simulation study considering comparative backtests of DieboldMariano type with a pointwise trafficlight illustration of Murphy diagrams. 
Date:  2019–07 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:1907.01306&r=all 
By:  Oliver Kley; Claudia Klüppelberg; Sandra Paterlini 
Abstract:  We introduce a statistical model for operational losses based on heavytailed distributions and bipartite graphs, which captures the event type and business line structure of operational risk data. The model explicitly takes into account the Pareto tails of losses and the heterogeneous dependence structures between them. We then derive estimators for individual as well as aggregated tail risk, measured in terms of ValueatRisk and ConditionalTailExpectation for very high confidence levels, and provide also an asymptotically full capital allocation method. Estimation methods for such tail risk measures and capital allocations are also proposed and tested on simulated data. Finally, by having access to realworld operational risk losses from the Italian banking system, we show that even with a small number of observations, the proposed estimation methods produce reliable estimates, and that quantifying dependence by means of the empirical network has a big impact on estimates at both individual and aggregate level, as well as for capital allocations. 
Keywords:  Bipartite graph, Estimation, Expected Shortfall, Extremal dependence, Operational Risk, Quantile risk measure, ValueatRisk 
Date:  2019 
URL:  http://d.repec.org/n?u=RePEc:trn:utwprg:2019/02&r=all 
By:  Rampini, Adriano A.; Viswanathan, S.; Vuillemey, Guillaume 
Abstract:  We study risk management in financial institutions using data on hedging of interest rate and foreign exchange risk. We find strong evidence that institutions with higher net worth hedge more, controlling for risk exposures, both across institutions and within institutions over time. For identification, we exploit net worth shocks resulting from loan losses due to drops in house prices. Institutions that sustain such shocks reduce hedging significantly relative to otherwise similar institutions. The reduction in hedging is differentially larger among institutions with high real estate exposure. The evidence is consistent with the theory that financial constraints impede both financing and hedging. 
Keywords:  Derivatives; Financial constraints; financial institutions; Foreign exchange risk; interest rate risk; Risk management 
JEL:  D92 E44 G21 G32 
Date:  2019–06 
URL:  http://d.repec.org/n?u=RePEc:cpr:ceprdp:13787&r=all 
By:  Maria Arduca; Pablo KochMedina; Cosimo Munari 
Abstract:  We describe a general approach to obtain dual representations for systemic risk measures of the "allocate first, then aggregate"type, which have recently received significant attention in the literature. Our method is based on the possibility to express this type of multivariate risk measures as special cases of risk measures with multiple eligible assets. This allows us to apply standard FenchelMoreau techniques to tackle duality also for systemic risk measures. The same approach can be also successfully employed to obtain an elementary proof of the dual representation of "first aggregate, then allocate"type systemic risk measures. As a final application, we apply our results to derive a simple proof of the dual representation of univariate utilitybased risk measures. 
Date:  2019–06 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:1906.10933&r=all 
By:  Pascal Traccucci; Luc Dumontier; Guillaume Garchery; Benjamin Jacot 
Abstract:  The quest for diversification has led to an increasing number of complex funds with a high number of strategies and nonlinear payoffs. The new generation of Alternative Risk Premia (ARP) funds are an example that has been very popular in recent years. For complex funds like these, a Reverse Stress Test (RST) is regarded by the industry and regulators as a better forwardlooking risk measure than a ValueatRisk (VaR). We present an Extended RST (ERST) triptych approach with three variables: level of plausibility, level of loss and scenario. In our approach, any two of these variables can be derived by providing the third as the input. We advocate and demonstrate that ERST is a powerful tool for both simple linear and complex portfolios and for both risk management as well as daytoday portfolio management decisions. An updated new version of the Levenberg  Marquardt optimization algorithm is introduced to derive ERST in certain complex cases. 
Date:  2019–06 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:1906.11186&r=all 
By:  Bräuning, Falk (Federal Reserve Bank of Boston); Fillat, Jose (Federal Reserve Bank of Boston) 
Abstract:  We use an expansive regulatory loanlevel dataset to analyze how the portfolios of the largest US banks have evolved since 2011. In particular, we analyze how the commercial and industrial and commercial real estate loan portfolios have changed in response to stresstesting requirements stipulated in the 2010 DoddFrank Act. We find that the largest US banks, which are subject to stress testing, have become more similar since the current form of the stress testing was implemented in 2011. We also find that banks with poor stress test results tend to adjust their portfolios in a way that makes them more similar to the portfolios of banks that performed well in the stress testing. In general, stress testing has resulted in more diversified bank portfolios in terms of sectoral and regional distributions. However, we also find that all the large US banks diversified in a similar way, creating a more concentrated systemic portfolio in the aggregate. 
Keywords:  bank correlations; concentration; portfolio similarity; stress tests; systemic risk 
JEL:  G21 G28 
Date:  2019–04–01 
URL:  http://d.repec.org/n?u=RePEc:fip:fedbcq:2019_001&r=all 
By:  Samuel Drapeau; Mekonnen Tadese 
Abstract:  Expectile bears some interesting properties in comparison to the industry wide expected shortfall in terms of assessment of tail risk. We study the relationship between expectile and expected shortfall using duality results and the link to optimized certainty equivalent. Lower and upper bounds of expectile are derived in terms of expected shortfall as well as a characterization of expectile in terms of expected shortfall. Further, we study the asymptotic behavior of expectile with respect to expected shortfall as the risk level goes to $0$ in terms of extreme value distributions. Illustrating the formulation of expectile in terms of expected shortfall, we also provide explicit or semiexplicit expressions of expectile for some classical distributions. 
Date:  2019–06 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:1906.09729&r=all 
By:  Triantafyllou, Athanasios; Dotsis, George; Sarris, Alexandros 
Abstract:  We empirically examine the predictability of the conditions which are associated with a higher probability of a price spike in agricultural commodity markets. We find that the forward spread is the most significant indicator of probable price jumps in maize, wheat and soybeans futures markets, a result which is in line with the “Theory of Storage”. We additionally show that some optionimplied variables add significant predictive power when added to the more standard information variable set. Overall, the estimated probabilities of large price increases from our probit models exhibit significant correlations with the historical sudden market upheavals in agricultural markets. 
Keywords:  Agricultural price spikes, Tail Risk Measure, Extreme Value Theory, Risk neutral moments, Agricultural Commodities, Basis, Theory of Storage 
Date:  2019–07–02 
URL:  http://d.repec.org/n?u=RePEc:esy:uefcwp:24921&r=all 
By:  Massimo Guidolin; Alexei Orlov 
Abstract:  We report systematic, outofsample evidence on the benefits to an already welldiversified investor that may derive from further diversification into various hedge fund strategies. We investigate dynamic strategic asset allocation decisions that take into account investors’ preferences as well as return predictability. Our results suggest that not all hedge fund strategies benefit a longterm investor who is already well diversified across stocks, government and corporate bonds, and REITs. Only strategies whose payoffs are highly nonlinear (e.g., fixed income relative value and convertible arbitrage), and therefore not easily replicable, constitute viable options. Most of the realized economic value fails to result from a meanvariance type of improvement but comes instead from an improvement in realized highermoment properties of optimal portfolios. Medium to highly riskaverse investors benefit the most from this alternative asset class. 
Keywords:  Strategic asset allocation, hedge fund strategies, predictive regressions, outofsample performance, certainty equivalent return. 
JEL:  G11 G17 G12 C53 
Date:  2018 
URL:  http://d.repec.org/n?u=RePEc:baf:cbafwp:cbafwp1890&r=all 
By:  Ernest Dautovic 
Abstract:  The paper evaluates the impact of macroprudential capital regulation on bank capital, risk taking behaviour, and solvency. The identication relies on an exogenous policy change in banklevel capital requirements across systemically important banks in Europe. A one percentage point hike in capital requirements leads to anaverage CET1 capital level increase of 13 percent improving their loss absorption capacity. 
Keywords:  capital requirements, risktaking, moral hazard, macroprudential policy 
JEL:  E51 G21 O52 
Date:  2019–06 
URL:  http://d.repec.org/n?u=RePEc:lau:crdeep:19.03&r=all 
By:  Marco Bee; Julien Hambuckers; Luca Trapin 
Abstract:  In this paper, we study the estimation of parameters for gandh distributions. These distributions find applications in modeling highly skewed and fattailed data, like extreme losses in the banking and insurance sector. We first introduce two estimation methods: a numerical maximum likelihood technique, and an indirect inference approach with a bootstrap weighting scheme. In a realistic simulation study, we show that indirect inference is computationally more efficient and provides better estimates in case of extreme features of the data. Empirical illustrations on insurance and operational losses illustrate these findings. 
Keywords:  Intractable likelihood, indirect inference, skewed distribution, tail modeling, bootstrap 
JEL:  C15 C46 C51 G22 
Date:  2019 
URL:  http://d.repec.org/n?u=RePEc:trn:utwprg:2019/11&r=all 
By:  Jennie Bai; Turan G. Bali; Quan Wen 
Abstract:  We propose a comprehensive measure of systematic risk for corporate bonds as a nonlinear function of robust risk factors and find a significantly positive link between systematic risk and the timeseries and crosssection of future bond returns. We also find a positive but insignificant relation between idiosyncratic risk and future bond returns, suggesting that institutional investors dominating the bond market hold welldiversified portfolios with a negligible exposure to bondspecific risk. The composite measure of systematic risk also predicts the distribution of future market returns, and the systematic risk factor earns a positive price of risk, consistent with Merton's (1973) ICAPM. 
JEL:  C13 G10 G11 G12 
Date:  2019–06 
URL:  http://d.repec.org/n?u=RePEc:nbr:nberwo:25995&r=all 
By:  Fabrizio Cipollini (Dipartimento di Statistica, Informatica, Applicazioni "G. Parenti", UniversitÃ di Firenze); Giampiero M. Gallo (Corte dei Conti and NYU in Florence, Italy); Edoardo Otranto (UniversitÃ di Messina, Italy) 
Abstract:  In this paper, we reconsider the issue of measurement errors affecting the estimates of a dynamic model for the conditional expectation of realized variance arguing that heteroskedasticity of such errors may be adequately represented with a multiplicative error model. Empirically we show that the significance of quarticity/quadratic terms capturing attenuation bias is very important within an HAR model, but is greatly diminished within an AMEM, and more so when regime specific dynamics account for a faster mean reversion when volatility is high. Model Confidence Sets confirm such robustness both inâ€“ and outâ€“ofâ€“sample. 
Keywords:  Realized volatility, Forecasting, Measurement errors, HAR, AMEM, Markov switching, Volatility of volatility 
JEL:  C22 C51 C53 C58 
Date:  2019–07 
URL:  http://d.repec.org/n?u=RePEc:fir:econom:wp2019_04&r=all 
By:  Martin Forde; Stefan Gerhold; Benjamin Smith 
Abstract:  We characterize the asymptotic smalltime and largetime implied volatility smile for the rough Heston model introduced by El Euch, Jaisson and Rosenbaum. We show that the asymptotic shortmaturity smile scales in qualitatively the same way as a general rough stochastic volatility model, and is characterized by the FenchelLegendre transform of the solution a Volterra integral equation (VIE). The solution of this VIE satisfies a spacetime scaling property which simplifies its computation. We corroborate our results numerically with Monte Carlo simulations. We also compute a power series in the logmoneyness variable for the asymptotic implied volatility, which yields tractable expressions for the vol skew and convexity, thus being useful for calibration purposes. We also derive formal asymptotics for the smalltime moderate deviations regime and a formal saddlepoint approximation for call options in the large deviations regime. This goes to higher order than previous works for rough models, and in particular captures the effect of the mean reversion term. In the large maturity case, the limiting asymptotic smile turns out to be the same as for the standard Heston model, for which there is a well known closedform formula in terms of the SVI parametrization. 
Date:  2019–06 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:1906.09034&r=all 
By:  Nick Whiteley 
Abstract:  This note outlines a method for clustering time series based on a statistical model in which volatility shifts at unobserved changepoints. The model accommodates some classical stylized features of returns and its relation to GARCH is discussed. Clustering is performed using a probability metric evaluated between posterior distributions of the most recent changepoint associated with each series. This implies series are grouped together at a given time if there is evidence the most recent shifts in their respective volatilities were coincident or closely timed. The clustering method is dynamic, in that groupings may be updated in an online manner as data arrive. Numerical results are given analyzing daily returns of constituents of the S&P 500. 
Date:  2019–06 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:1906.10372&r=all 