Shihao Gu
Bryan Kelly
Dacheng Xiu
methods：
item is asset (one asset level)
1 predictors –> return –> portfolio selection
2 predictors –> covariance –> portfolio selection
Can Machine Learning-Based Portfolios Outperform Traditional
Risk-Based Portfolios?
3 predictors –> return covariance, skewness… –> portfolio selection
item is portfolio (portfolios level)
4 preditors –> portfolios’ return –> portfolio selection
5 predictors –> portfolios’ attributes –> portfolio selection
