Paper Draft — Portfolio VaR via Eigenvalue-Conditioned Lognormal Mixtures
Unreviewed draft. This paper has not been human-reviewed.
Mathematical claims may be unverified. Use with appropriate caution.
Summary
We present a method for computing Value-at-Risk of portfolios whose assets
follow correlated geometric Brownian motion. The portfolio value is a weighted
sum of correlated lognormals — the Fenton Distribution — whose CDF has no
closed form.
Length
4,114 words
Status
Draft