- Use total, upper, down, relative volatility factors to find Alpha.
- Define own functions for trading direction & volume.
- Implement whole trading process & back-test with daily frequency.
- Calculate and record important financial data.
- Plot charts & Output trading financial data into files for visualization.
- Volatility factors:
- Total Volatility (VOL):
It measures the total volatility of the instrument. With higher volatility, it demands for higher return. - Upper/Down Volatility (UVOL/DVOL):
It measures the volatility by trend (go up or down) of the instrument. With higher unilateral volatility, the price volatility is higher in this side. - Relative Volatility (RSJ): 𝐑𝐒𝐉 = 𝐔𝐕𝐎𝐋 − 𝐃𝐕𝐎𝐋
It measures the relative volatility of the instrument. With higher absolute value of relative volatility, it shows short-term pressure (sharp long and steady short power), and lower future return.
- Financial data: Holding, Position, Total assets, Available cash, Profit and Loss, Return, Sharp's retio, volatility...
- Trading records: Trading direction, Trading volume, Turnover rate, Trading times, Maximum drawdown...
- RSJ has the best result in most years.
- Volatility Factors Windows: Shorter Time is better. 5 days, 10 days, 15 days windows perform better than long-time windows.
- Macro Factors Windows: Longer Time is better. 24 months provides a better PnL, return and maximum drawdown.
- Period Analysis:
- This model follows the trend of the market.
- It has a conservative approach. When the market is very volatile, it does not go as high or as low as CSI500.
- It needs improvement in a market with high volatility.
- It performs good in stable conditions. It overperforms the Chinese market by 30% in last two years.
Data: CSI500 Index
Data Source: Wind