Nematrian
/


Market Risk [21]

Go to: Summary | Previous | Next   
Bullet points include: Given n time series observations: x1, x2, ..., xn, Order realisations and re-label so x(1) <= x(2) ... <= x(n). Then x(r) is the r-th order statistic, where r integer between 1 and n. Obvious non-parametric estimator for VaR is k-th order statistic where: Implicitly assumes no temporal dependence (e.g. no GARCH effects). VaR more natural non-parametric risk measure than tracking error / volatility

NAVIGATION LINKS
Contents | Prev | Next | ERM Lecture Series

Desktop view | Switch to Mobile