Nematrian
/

Extreme Events and Portfolio Construction [29]

Go to: Summary | Previous | Next   
Bullet points include: Most important (predictable) single contributor to fat tails seems to be time-varying volatility. So: Calculate covariance matrix between return series after stripping out effect of time-varying volatility? Optimise as you think fit (standard, “robust”, Bayesian, BL, ...), using adjusted covariance matrix. Adjust risk aversion/risk budget appropriately and then unravel time-varying volatility adjustment. Or reverse optimise using implied alphas derived from adjusted covariance matrix. Implicitly assumes all adjusted return series ‘equally’ fat-tailed. Capturing other sources of fat-tailed behaviour adds more complexity

NAVIGATION LINKS
Contents | Prev | Next | Library

Desktop view | Switch to Mobile