Abstract
The purpose of this paper is to evaluate the quantitative standard d. laid down under the second and third Basel Accords for the implementation of internal market risk models by banks. This standard specifies a minimum historical observation period for VaR models. The paper finds a shortage of research that studies the impact of the quantitative standard on choice of VaR methods and their outputs. The specification of a minimum observation period of one year results in a smoother and less responsive VaR estimate and rules out conditional volatility models. The paper examines the assumption that longer historical periods result in more reliable VaRs and are more likely to incorporate stressed episodes. Both assumptions are questionable.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call