High Watermarks of Market Risk

Extreme Volatilities, Financial Crises and L-Moment Estimations of Tail-Indexes

“…As previously shown in Maillet et al (2010), a final reality check, reported in Table 7, shows inadequacies of some of the simplest measures of risk, since they either greatly underestimate or largely overestimate the Return-times for the peak events observed in the sample. The use of the Tail-index estimates, relative to the Realized Volatilities, however, produces some more reasonablelooking values for the Return-times of the high watermarks of volatility.…”

“…Following Bali and Weinbaum (2005) and Maillet et al (2010), the aim of the present article is to, furthermore, study the main high and low-frequency measures of volatility -focusing on extreme events, in order to find if there are glaring discrepancies between the empirical evidence and the usual assumptions about the distribution of volatility. The distribution will be estimated using various techniques including different L-moment methods.…”

“…

Following Bali and Weinbaum (2005) and Maillet et al (2010), we present several estimates of volatilities computed with high-and low-frequency data and complement their results using additional measures of risk and several alternative methods for Tail-index estimation. The aim here is to confirm previous results regarding the slope of the tail of various risk measure distributions, in order to define the high watermarks of market risks.

…”

“…However, first preliminary tests were based on direct 30" quotes, then ran another time on a 5' aggregation scheme, whilstMaillet et al (2010) used a 20' sampling rate, with no resulting change in the overall qualitative following results. Additional measures -not reported here for space reason but available on demand -have been also computed for 60' and 120' sampling rates, and lead once again to the same conclusion.…”