A Synthetic Regression Model for Large Portfolio Allocation

Portfolio allocation is an important topic in financial data analysis. In this article, based on the mean-variance optimization principle, we propose a synthetic regression model for construction of portfolio allocation, and easy to implement approach generate sample model. Compared with existing literature proposed method generating provides more accurate approximation response variable when number assets under consideration large. Due embedded leave-one-out idea, generated by has weaker within correlation, which makes resulting close optimal one. This intuitive conclusion theoretically confirmed be true asymptotic properties established article. We have also conducted intensive simulation studies article compare ones, found works better. Finally, apply real datasets. The yielded returns look very encouraging.