2012Fall7646 Homework 7
In this homework you will take the output of your Event Study work to build a real trading algorithm. Specifically, you should choose an event from the ones you have experienced in this class, assess it and tune it using the Event Profiler, then back test it in the following manner:
1. Have your program output a series of trades based on events. Instead of putting a 1 in the event matrix, output to a file
Date, BUY, AAPL, 100 Date + 5 days, SELL, AAPL, 100
2. Feed that file into your market simulator.
3. Report the performance of your strategy in terms of total return and Sharpe Ratio for the time period.
Experiments to Run
You should run two experiments.
- Experiment 1: Use the actual close $5.00 event with the 2012 SP500 data. I expect that this strategy should make money. We will use this example to make sure everybody is getting "correct" answers. You can share your numerical results with others to make sure you're on the right track. Use the following parameters:
- Starting cash: $50,000
- Start date: 1 January 2008
- End date: 31 December 2009
- When an event occurs, buy 100 shares of the equity on that day.
- Sell automatically 5 days later.
- Experiment 2: Devise your own event and trading strategy.
For both experiments, you should:
- Create a chart to illustrate the performance of your fund.
- Compute the Sharpe Ratio, total return and STDDEV of daily returns.
Submit these files:
- strategy.py Python program that creates an event study AND generates trades that you feed to the simulator you created using YOUR strategy.
- report.pdf a PDF including the Event Study, your performance chart, along with text of your report that describes each strategy and describes its performance. Include metrics of performance for you strategy including Sharpe Ratio, total return, STDEV of return.
How to submit
Go to the t-square site for the class, then click on the "assignments" tab. Click on "add attachment" to add your files. Once you are sure you've added the files, click "submit."