Funds that invest in illiquid assets report returns with spurious autocorrelation. Conse-quently, investors need to unsmooth returns when evaluating the risk exposures of these funds. We show that funds investing in similar assets have a common source of spuri-ous autocorrelation, which is not addressed by commonly-used unsmoothing methods, leading to underestimation of systematic risk. To address this issue, we propose a gener-alization of these unsmoothing techniques and apply it to hedge funds and commercial real estate funds. Our empirical results indicate our method signiﬁcantly improves the measurement of risk exposures and risk-adjusted performance, with stronger results for more illiquid funds.