This issue of The Journal of Risk addresses liquidity risk and model risk using the same methodology: namely, extreme value theory. In addition, model risk is studied from an empirical perspective and a new type of bond for insurance companies is analyzed.
Excessive redemption is one of the main sources of liquidity risk in open-end mutual funds. In our first paper, "Modeling redemption risks of mutual funds using extreme value theory", Sascha Desmettre and Matthias Deege adapt the liquidity-at-risk concept, originally developed for the banking sector, to mutual funds investing. Specifically, they show how to integrate mutual fund data into a model for which they develop a practical procedure to determine the threshold parameter of a generalized Pareto distribution, and they calibrate it with backtesting.
In the issue's second paper, "The role of model risk in extreme value theory for capital adequacy", Ralf Kellner, Daniel Rösch and Harald Scheule focus on the sensitivity of risk measures, assessed based on extreme value theory. In particular, they highlight the difference between first-order estimation effects and second-order dispersion effects, with the latter having a greater impact on regulatory capital requirements.
Model risk is also addressed empirically in our third paper: "Relative performance persistence of financial forecasting models and its economic implications" by Eduard Baitinger, Christian Fieberg and Thorsten Poddig. In contrast to macroeconomic parameter estimation, the authors show that models for financial portfolio management exhibit little consistency in their performance over time. Because of reversals between best and worst models, they recommend using models that perform in the middle of the pack.
In the fourth and final paper of the issue, "The valuation of contingent convertible catastrophe debt under simple solvency and liquidity covenants", Nick Georgiopoulos studies a bond with a binary payoff that is unique to insurance companies: the bond is written off, with no bankruptcy implication, if specific, and usually rare, events occur. The author's main focus is on capital structure for insurers and the trade-off that needs to be contemplated between equity and this type of bond in the presence of insurance claims.
University of Florida
This paper studies the impact of model risk on EVT methods when determining the value-at-risk and expected shortfall.
This paper addresses the issue of model selection risk by examining whether a model’s past performance in forecasting expected returns provides an indication of its future forecasting performance.
The valuation of contingent convertible catastrophe debt under simple solvency and liquidity covenants
This paper studies a new write-off debt instrument (called CoCoCAT bond) whose writeoff is triggered by solvency and event-driven covenants.
This paper shows how redemption risks of mutual funds can be modeled using the peaks-over-threshold approach from extreme value theory.