A summary for what have been done in the prism project:
- Basic time series routines: normalization, distance, linear fit, local max/min, data reduction by turning points, piecewise, etc..
- A very versatile pattern searching algorithm.
- Basic OHLC data processing routines, split restore, shrink by week, by month, etc.
- Centralized OHLC data storage. OHLC data are stored in key-value database kyoto cabinet.
- A extensible indicator calculation framework, and implemented following indicator: SMA, EMA, MACD, RSI, KDJ, CR(new), etc.
- A rule-based strategy and the back test framework for it.