We address the problem of poor portfolio performance when a minimum-variance portfolio is constructed using the sample estimates. Estimation errors are mostly blamed for the poor portfolio performance. However, we argue that even small unbiased estimation errors can lead to significantly bad...

In this paper, I reinvestigate the popular belief that the Maximum-Sharpe (MS) portfolio has terrible out-of-sample performance because of estimation errors in the expected return. I show that even without estimation error, the MS portfolio could still perform terribly under a measurement that...

We propose a new dimension reduction procedure for portfolio optimization. The vanilla principal component analysis (PCA) restricts the portfolio on the linear subspace spanned by the PCs, which often requires many PCs to performance well. In our framework, the linear subspace is based on a most...