Extreme Value Theory for Finance: A Survey

Abstract

Extreme value theory is concerned with the study of the asymptotical distribution of extreme events, that is to say events which are rare in frequency and huge with respect to the majority of observations. Statistical methods derived from this theory have been increasingly employed in finance, especially in the context of risk measurement. The aim of the present study is twofold. The first part delivers a critical review of the theoretical underpinnings of extreme value theory. The second part provides a survey of some major applications of extreme value theory to finance, namely its use to test different distributional assumptions for the data, Value-at-Risk and Expected Shortfall calculations, asset allocation under safetyfirst type constraints and the study of contagion and dependence across markets under stress conditions.

Enhancing Anomaly Detection for Cultural Heritage via Long Short-Term Memory with Attention Mechanism

“…EVT is commonly used to predict the probability of extreme events, and one important tool is the generalized extreme value distribution. It extensively involves hydrology [42,43], meteorology [44], finance [45,46], and is calculated through Equation (13).…”

Designing energy-efficient serial production lines: The unpaced synchronous line-balancing problem

“…There are numerous applications of extreme value theory in energy management (e.g., Chan & Nadarajah, 2015,Feng, Wei, & Wang, 2012, engineering (Castillo, Hadi, Balakrishnan, & Sarabia, 2005), and finance (Rocco, 2014); however, very little has been done in the management domain (Buam & McKelvey, 2006).…”

Portfolio Tail Risk: A Multivariate Extreme Value Theory Approach

“…The importance of tail behavior in risk assessment has been emphasized by many authors, such as Poon et al [ 1 ], Rodriguez [ 2 ], Ning [ 3 ], Okimoto [ 4 ], Ning [ 5 ], Christoffersen et al [ 6 ], Rocco [ 7 ], Gkillas et al [ 8 ], and Gkillas et al [ 9 ], and many distributional assumptions have been introduced so far to improve risk management and asset pricing practices (see, for instance, [ 10 , 11 ]). Perhaps the most obvious alternative to the usual parametric approaches is the Extreme Value Theory (EVT) since it concentrates on extreme losses.…”