Financial Data Warehouse

High Level Logical Flow:

1. Entry Point Class
   • collects/reads/encapsulates all the settings/configurations for a single BackTest instance
   • output: instance of Back Test Config Class
2. Back Test Config Class
   • Stores all the settings/parameters/configurations required by the Back Test Class
   • Start Date, End Date, Stock List,...?
3. Data Getter Class
   • input: instance of Back Test Class
   • Use info from input to know what data is required then queries some data source (csv file, database, internet, whatever)
   • output: instance(s) Price Series Class
4. Price Series Class
   • Maybe just a Pandas array, maybe this isn't needed
5. Crystal Ball Class
   • input: instance of Price Series Class
   • generates buy or don't buy signals for each stock for each day
   • output: instance(s) Buy Signal Series
6. Buy Signal Series
   • Maybe just a Panada array, maybe this isn't needed
7. Balancer Class
   • input: instance(s) Buy Signal Series
   • implements daily re-balancing of holdings according input
   • output: instance Portfolio Class
8. Portfolio Class
   • all sorts of analytics
     • overall return calculation
     • volatility
     • Sharpe Ratio
     • daily return calculations

Stock Trading Strategy Backtesting

Trade Signal

• symbol
• date

Trade Signal Collection

Back Test Bot object

Member variables:

1. sell every thing we bought yesterday at today's prices
   1. determine % change in stock price of all stocks bought yesterday
   2. determine % change of holding
      1. yesterdays allocation * % change of stock price
2. determine which stocks are a buy at today's prices
   1. for each stock crystal ball says whether or not to buy
3. determine today's allocation for the buys
   1. 1 / count of buys signals
4. hold stocks. go to tomorrow.