Description:

tradingsystem 0.0.1

TradingSystem
An event-driven backtest/realtime quantitative trading system.
Architecture
Data Handler

Handle live and historical OHLCV data
Generate a BarEvent which contains ticker symbol and OHLCV data
Strategy
Get BarEvents and make trading decisions based on predefined rules
Generate a SignalEvent which contains order type (market/limit order) and the direction (long/short)
Order Handler
Get SignalEvents and determine the quantities that should be bought/sold
Generate an OrderEvent containing order type, direction, and quantities
Broker
Get OrderEvents and route orders to a simulated or real brokerage
Once orders are filled, it creates a FillEvent which has executed price, commission, and the exchange where the order was filled
Portfolio
Manage the whole portfolio. It tracks PnL, position, and the average price of ticker symbols
Get BarEvents and FillEvents to update the portfolio

Installation

create a virtual environment
pip install tradingsystem

Examples
For backtesting,
import queue
from tradingsystem.data_handler.historical_bar_handler import HistoricalBarHandler, get_data_from_csv, get_data_from_yahoo_finance
from tradingsystem.strategy.sma_crossover import SMACrossover
from tradingsystem.portfolio.portfolio import Portfolio
from tradingsystem.order_handler.max_order_handler import MaxOrderHandler
from tradingsystem.broker.simulated_broker import IBSimulatedBroker
from tradingsystem.statistics.statistics import Statistics
from tradingsystem.common import SessionType
from tradingsystem.simulated_trading_session import TradingSession

events_queue = queue.Queue()
init_asset_val = 10000
tickers_data = {}
amzn = get_data_from_yahoo_finance("AMZN", "2022-01-01", "2022-05-01", "1d")
ticker_list = ['AMZN']
tickers_data['AMZN'] = amzn
historical_bar_handler = HistoricalBarHandler(tickers_data, events_queue)
portfolio = Portfolio(init_asset_val)
order_handler = MaxOrderHandler(portfolio, events_queue, SessionType.BACKTEST)
sma_crossover = SMACrossover(ticker_list, 10, 20, events_queue, SessionType.BACKTEST, portfolio)
simulated_broker = IBSimulatedBroker(events_queue, historical_bar_handler)
stat = Statistics(init_asset_val)
trading_session = TradingSession(historical_bar_handler, sma_crossover, portfolio,
order_handler ,simulated_broker, events_queue, stat)
#start backtest
trading_session.start_trading()

For live trading in IB,

Open Trader Workstation (TWS) or IB Gateway
Change timezone to UTC in options
Run the code below

import queue
from tradingsystem.data_handler.ib_real_time import IBRealTimeBarHandler
from tradingsystem.strategy.sma_crossover import SMACrossover
from tradingsystem.order_handler.max_order_handler import MaxOrderHandler
from tradingsystem.portfolio.portfolio import Portfolio
from tradingsystem.broker.ib_broker import IBBroker
from tradingsystem.statistics.statistics import Statistics
from tradingsystem.ibtws.twsclient import TWSClient
from tradingsystem.common import SessionType
from tradingsystem.trading_schedule.fx_schedule import FXSchedule
from datetime import datetime
from tradingsystem.ib_trading_session import TradingSession

#set up
events_queue = queue.Queue()
init_asset_val = 10000 #in USD
session_type = SessionType.LIVE
twsclient = TWSClient("127.0.0.1", 7497, 0)
trading_schedule = FXSchedule(2022)
symbol_list = ['EUR/USD']
ib_bar_handler = IBRealTimeBarHandler(twsclient, symbol_list, 300,
"MIDPOINT", True, events_queue,
)
portfolio = Portfolio(init_asset_val)
max_order_handler = MaxOrderHandler(portfolio, events_queue, session_type)
sma_crossover = SMACrossover(symbol_list, 10, 20, events_queue, session_type, portfolio)
ib_broker = IBBroker(twsclient, events_queue, symbol_list)
stat = Statistics(init_asset_val)
trading_end = datetime(2022,12,31, 0, 00)
trading_session = TradingSession(twsclient, ib_bar_handler, sma_crossover, portfolio,
max_order_handler, ib_broker, events_queue,
trading_schedule, trading_end, stat)
#start trading
trading_session.start_trading()

License

For personal and professional use. You cannot resell or redistribute these repositories in their original state.

Share

• Share on Facebook
• Share on Twitter