The whole project contains two part: Pairs trading strategy & Technical Indicators strategy. Codes presented here are for technical indicators strategy, and are solely developed by myself.

------------------------------------------------------------- CODES -----------------------------------------------------------------

• StockDataCollection: Collect data from yahoo finance.

• Quant Trading:

  A simplified version of codes that display the core logic of our strategy. Further explained below.

------------------------------------------------------------- NOTES ----------------------------------------------------------------- *Techinical Indicators:

To capture trading signals as accurate and comprehensive as possible, a variety of empirical technical indicators with different focuses and purposes have been selected for the compilation of this strategy, as shown below:

• Trend Indicators (MACD, Parabolic SAR) are selected to eliminate the noise of daily volatility
• Momentum Indicators (KDJ, RSI, ICH) are deployed to show the fluctuation of a stock
• Volatility Indicators (BBANDS, Illiquidity, ATR) are used as a protection mechanism against breakout
• Volume Indicators (OBV, ROC) would provide indication on the strength and direction of price movement.

*Key Steps:

1. Calculation: With previous market data, we calculate the result from each individual indicator respectively and from the combination of different indicators.

2. Encoding: The observed range and trends from the above mentioned process then yield us new trading signals indicating long(1), weak long(0.5), hold(0), weak short(-0.5), and short(-1) respectively.

3. Integration & Prediction: The encoded signals are combined together to produce an integrated final signal for our trade under specific strategy. Four models are applied: Random Forest, Neural Networks, SVM and Logistic Regressions, while predictions made by logistic regressions are prioritized based on backtesting results.