Comparing Different Methods of Calculating Value at Risk

16 Pages Posted: 25 Apr 2005

See all articles by Tapen Sinha

Tapen Sinha

Chennai Mathematical Institute; Nottingham University Business School; Instituto Tecnológico Autónomo de México (ITAM) - Division of Actuarial Science, Statistics and Mathematics

Francisco Chamu

University of North Carolina at Chapel Hill - Department of Statistics

Date Written: April 2000

Abstract

We compare the performance of three different methods of calculating VaR in the context of volatile markets. Some of these methods are routinely used for banks, pension funds and mutual funds without critical valuation of their efficiency. We examine weaknesses of these methods by using five different tests. They are (1) test based on the time until first failure, (2) test based on failure rate, (3) test based on expected value, (4) test based on autocorrelation, and (5) test based on (rolling) mean absolute percentage error.

Keywords: Value at risk, first failure, failure rate, expected value, autocorrelation, mean absolute percentage error

JEL Classification: G1, G11, G18

Suggested Citation

Sinha, Tapen and Chamu, Francisco, Comparing Different Methods of Calculating Value at Risk (April 2000). Available at SSRN: https://ssrn.com/abstract=706582 or http://dx.doi.org/10.2139/ssrn.706582

Tapen Sinha (Contact Author)

Chennai Mathematical Institute ( email )

SIPCOT
Chennai, 600017
India

Nottingham University Business School ( email )

Jubilee Campus
Wollaton Road
Nottingham, NG8 1BB
United Kingdom

Instituto Tecnológico Autónomo de México (ITAM) - Division of Actuarial Science, Statistics and Mathematics ( email )

Rio Hondo #1 Col. Tizapan San Angel
C.P. 01000 Del. Alvaro Obregon
52 55 5628 4088 (Phone)
52 55 5628 4086 (Fax)

Francisco Chamu

University of North Carolina at Chapel Hill - Department of Statistics ( email )

102 Ridge Road
Chapel Hill, NC NC 27514
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
2,445
Abstract Views
9,347
Rank
10,915
PlumX Metrics