Abstract :
[en] Extreme value regression (EVR) offers a convenient framework to assess the effect
of market variables on hedge funds tail risks. However, its major limitation lies in
the need to select a threshold below which data are discarded, leading to significant
estimation inefficiencies. Our main contribution consists therefore in introducing a
method to estimate simultaneously the tail and the threshold parameters from the
entire sample, improving estimation efficiency. To do so, we extend the tail regression
model to non-tail observations with an auxiliary splicing density, enabling the thresh-
old to be internally determined by the tail parameters. We then apply an artificial
censoring mechanism of the likelihood contributions to decrease specification issues
at the estimation stage. We illustrate the superiority of our approach for inference
over classical peaks-over-threshold methods in a simulation study. Empirically, we
investigate the determinants of Long/Short Equity hedge funds tail risks over time
with our method, using pooled returns of 1,484 hedge funds. We find a significant link
between tail risks and factors such as liquidity indicators. Moreover, sorting funds
along exposure to our tail risk measure discriminates between high and low alpha
funds, supporting the existence of a fear premium
