Warrington College of Business, University of Florida
Equity volatility estimation that incorporates macroeconomic data sampled at lower frequency is addressed in this issue of The Journal of Risk, as are modeling alternatives to the benchmark floating interest rate. Additionally, the role of tail distributions is explored in the context of a graph-theoretic approach to systemic risk and in quantile-based distortion risk measures.
In the issue’s first paper, “Forecasting stock market volatility: an asymmetric conditional autoregressive range mixed data sampling (ACARR-MIDAS) model”, Xinyu Wu, Yang Han and Chaoqun Ma propose a model that captures conditional volatility asymmetry and volatility persistence using range-based estimates and incorporating macroeconomic data through the mixed data sampling (MIDAS) approach. The authors show that their model, which decomposes the conditional range into short and long-term components, is competitive relative to alternatives when it comes to both its forecast accuracy and its robustness to window selection.
Our second paper, “The impact of compounding on bond pricing with alternative reference rates” by Dario Cziráky and Ana Ponikvar, addresses the modeling impact of alternative reference rates, such as the UK Sterling Overnight Index Average (SONIA) and the US Secured Overnight Financing Rate (SOFR), on bond pricing. Using a standard mean-reverting Gaussian short-rate model – instead of a more realistic but less tractable one that captures stochastic volatility and jumps that are characteristic of such rates – the authors find that their model leads to negligible bias for tenors up to one year, with a moderate increase beyond one year.
In “Time-varying tail dependence networks of financial institutions”, the third paper in this issue, Fenghua Wen, Kaiyan Weng and Jie Cao study the systemic risk of the Chinese financial system. Using an approach based on a time-varying Clayton copula and the minimum spanning tree algorithm to account for lower-tail dependence and network topology, the authors are able to capture recent financial crises, including China’s “interbank money shortage” of 2013 and the market crash of 2015, and identify seven systemically important financial institutions.
In the issue’s fourth and final paper, “An examination of the tail contribution to distortion risk measures”, Miguel Santolino, Jaume Belles-Sampera, José María Sarabia and Montserrat Guillen propose an approach to assess the tail subadditivity of quantile-based distortion measures, such as value-at-risk and tail value-at-risk. Their approach results in an expression that helps identify the contribution of extreme losses to the reported risk value, and thus the diversification effect that is linked to the tails of the loss distributions.
Forecasting stock market volatility: an asymmetric conditional autoregressive range mixed data sampling (ACARR-MIDAS) model
This paper proposes an extension of the classical CARR model, the ACARR-MIDAS model, to model volatility and capture the volatility asymmetry as well as volatility persistence.
This paper looks at the impact of compounding on zero-coupon bond prices by considering the short rate when it follows a Gaussian diffusion process or a stochastic volatility jump-diffusion process.
In this paper time-varying tail dependence networks are constructed to investigate the complex interdependencies in the financial system.
This paper reports a method for analyzing the influence of the tail in calculations of distortion risk measures.