| 简介： | We propose a novel procedure to estimate the Minimum Spanning Set (MSS), which is the smallest subset of risky assets that spans the mean-variance efficient frontier of the full asset universe. We derive sufficient conditions for identification and establish the asymptotic properties of the estimator. Extensive Monte Carlo simulations demonstrate strong finite-sample performance. Applying our method to asset pricing data, we uncover the critical role of factor momentum, along with individual stock momentum and a set of established risk factors, in shaping the mean-variance frontier, and we further identify the underlying sources of their contributions. |