Portfolio optimization using asymmetry robust mean absolute deviation model

- We construct an asymmetry robust mean absolute deviation (ARMAD) model that takes the asymmetry distribution of returns into consideration.
- We test different robust strategies using the historical data of Chinese small cap stocks based on the growing and declining market, respectively.
- Computational experiments show that the ARMAD method can distinguish the high return stocks.

In this paper, we construct an asymmetry robust mean absolute deviation (ARMAD) model that takes the asymmetry distribution of returns into consideration. We test different robust strategies using the historical data of Chinese small cap stocks based on the growing and declining market, respectively. Computational experiments show that the ARMAD method can distinguish the high return stocks. Since there is short-run persistence of relative performance of the stocks, the portfolios constructed by the ARMAD model can provide investors with good guidance in the near future.