Abstract
Value-at-Risk (VaR) is a well-accepted risk metric in modern quantitative risk management (QRM). The classical Monte Carlo simulation (MCS) approach, denoted henceforth as the classical approach, assumes the independence of loss severity and loss frequency. In practice, this assumption does not always hold true. Through mathematical analyses, we show that the classical approach is prone to significant biases when the independence assumption is violated. This is also corroborated by studying both simulated and real-world datasets. To overcome the limitations and to more accurately estimate VaR, we develop and implement the following two approaches for VaR estimation: the data-driven partitioning of frequency and severity (DPFS) using clustering analysis, and copula-based parametric modeling of frequency and severity (CPFS). These two approaches are verified using simulation experiments on synthetic data and validated on five publicly available datasets from diverse domains; namely, the financial indices data of Standard & Poor’s 500 and the Dow Jones industrial average, chemical loss spills as tracked by the US Coast Guard, Australian automobile accidents, and US hurricane losses. The classical approach estimates VaR inaccurately for 80% of the simulated data sets and for 60% of the real-world data sets studied in this work. Both the DPFS and the CPFS methodologies attain VaR estimates within 99% bootstrap confidence interval bounds for both simulated and real-world data. We provide a process flowchart for risk practitioners describing the steps for using the DPFS versus the CPFS methodology for VaR estimation in real-world loss datasets.
Highlights
Research activities in quantitative risk management (QRM) have been steadily growing, due to its capability to analyze, quantify, and mitigate risks associated with various events that cause losses
Suppose that the loss severity consists of daily loss data, while the loss frequency consists of counts of the number of losses that occurs in a month
We describe results for two sample cases and show how the driven partitioning of frequency and severity (DPFS) approach splits the loss frequency and loss severity data
Summary
Research activities in quantitative risk management (QRM) have been steadily growing, due to its capability to analyze, quantify, and mitigate risks associated with various events that cause losses. Robust VaR measures have been developed to estimate the predictive distributions of generalized auto-regressive conditional heteroskedasticity (GARCH) models (Mancini and Trojani 2011) This method uses a semiparametric bootstrap to obtain robust VaR predictions in the market risk contexts in which a GARCH model is applicable. The operational risk framework uses a generalized approach of looking at the statistical characteristics of the losses based on their magnitude (severity) and the number of occurrences (frequency). This joint severity-frequency model is defined following the guidelines from the Basel committee on banking supervision (BCBS) (Balthazar 2006). The aggregated loss, S, is modeled as a random sum in the following manner: i.i.d
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have