EVT and alternatives [20]

Go to: Summary | Previous | Next   
Bullet points include: Model with a mixture of multivariate Normal distributions (or GPDs, …). Time-stationary? Maybe not realistic? Time-varying? (Discrete) regime switching, and/or (Continuous) parameterisation (and continuous time?). However: Even a mixture of just two multivariate Normal distributions involves estimation of twice as many parameters. Results even more sensitive to input assumptions. Time varying => dynamic => sensitivity to transaction costs

Contents | Prev | Next | Library

Desktop view | Switch to Mobile