Abstract

The tail of equity returns is typically governed by a power law, however, the constancy of the so-called tail index α, which dictates the tail decay, has been hardly studied. Using the Hill estimator to estimate the tail index, we study the finite sample properties of endogenous stability tests for α. We show that the finite sample critical values strongly depend on the underlying distributional assumptions for the stock returns. We therefore recommend a bootstrap-based version of the stability test as an alternative to the test’s asymptotic distribution. Upon applying this stability test to return tails of developed and emerging equity markets, the evidence for structural shifts is found to be rather weak. This is reassuring news for the proponents of Extreme value Theory (EVT). They typically assume stationary tail behavior when applying tail index and extreme quantile estimators to the downside risk of equity portfolios.

Full Text
Paper version not known

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