For assets that are sparsely traded, the time-variance of prices is of limited interest. It is more meaningful to compute price variance across assets and use that as an indicator of investment risk.

On a large database of funds invested in illiquid assets such as real estate, private equity and infrastructure, the cross-sectional variance -or dispersion- of fund performances is measured within and between the asset classes. The covariance structure obtained in this way has similar features as one between regularly-traded assets that is computed over time.

The author gives demonstration that the cross-sectional covariance between illiquid assets gives a workable framework for analysing risk and constructing portfolios with the same methods that are habitually deployed for liquid assets.