While there are no significant investment characteristics that inhibit art from being considered as an asset, a major hurdle has long been the lack of a systematic measure of its financial performance. Due to its heterogeneity (each piece is different) and its infrequency of trading (the exact same piece does not come to the market very often), the determination of changes in market value is difficult to ascertain. Two estimation methods are commonly used to construct indices. Repeat-sales regression (RSR) uses prices of individual objects traded at two distinct moments in time. If the characteristics of an object do not change (which is usually so for collectibles), the heterogeneity issue is bypassed. The basic idea of the hedonic regression (HR) method is to regress prices on various attributes of objects (dimensions, artist, subject matter, etc.) and to use the residuals of the regression which can be considered as "characteristic-free prices" to compute the price index. The chapter deals with the basics of hedonic and repeat-sales estimators, and tries to interpret in economic terms what both are trying to achieve. It also goes into some more technical details which may be useful for researchers who want to construct such indices, and gives some guidelines on how to go about collecting data, and the choice between RSR and HR that this induces. Both methods are compared using simulated returns, pointing to which method should be used given the data at hand.