Data Feeds

Data feeds provide market data to the backtesting engine.

Data Feeds Module - Data source implementations.

This module provides data feed implementations for various data sources including CSV files, pandas DataFrames, Yahoo Finance, Quandl, InfluxDB, and the unified bt_api_py live interface.

Data Feed Types:

  • CSV Feeds: GenericCSVData, BTCSV, MT4CSV, SierraChart, VChartCSV

  • Pandas: PandasData for pandas DataFrame integration

  • Online: Yahoo Finance, Quandl, BtApiFeed

  • Utilities: Chainer, RollOver for data manipulation

Example

Loading data from a CSV file: >>> data = bt.feeds.GenericCSVData( … dataname=’data.csv’, … datetime=0, … open=1, … high=2, … low=3, … close=4, … volume=5 … ) >>> cerebro.adddata(data)

Overview

Data feeds are the source of market data for strategies. Backtrader supports various data formats and sources.

Built-in Data Feeds

CSV Data

  • GenericCSVData: Flexible CSV reader

  • YahooFinanceCSVData: Yahoo Finance format

  • BacktraderCSVData: Backtrader native format

Pandas

  • PandasData: Load from pandas DataFrame

  • PandasDirectData: Direct DataFrame access

Live Data

  • IBData: Interactive Brokers

  • OandaData: OANDA forex

  • CCXTData: Cryptocurrency exchanges

Using CSV Data

# Generic CSV
data = bt.feeds.GenericCSVData(
    dataname='data.csv',
    dtformat='%Y-%m-%d',
    datetime=0,
    open=1,
    high=2,
    low=3,
    close=4,
    volume=5,
    openinterest=-1
)

# Yahoo Finance CSV
data = bt.feeds.YahooFinanceCSVData(
    dataname='AAPL.csv',
    fromdate=datetime(2020, 1, 1),
    todate=datetime(2021, 1, 1)
)

Using Pandas Data

import pandas as pd

df = pd.read_csv('data.csv', parse_dates=['date'])
df.set_index('date', inplace=True)

data = bt.feeds.PandasData(
    dataname=df,
    datetime=None,  # Use index as datetime
    open='open',
    high='high',
    low='low',
    close='close',
    volume='volume'
)

Resampling Data

# Load minute data
data = bt.feeds.GenericCSVData(dataname='minute_data.csv')
cerebro.adddata(data)

# Resample to hourly
cerebro.resampledata(
    data,
    timeframe=bt.TimeFrame.Minutes,
    compression=60
)

# Resample to daily
cerebro.resampledata(
    data,
    timeframe=bt.TimeFrame.Days,
    compression=1
)

Multiple Data Feeds

# Add multiple stocks
for symbol in ['AAPL', 'GOOGL', 'MSFT']:
    data = bt.feeds.YahooFinanceCSVData(
        dataname=f'{symbol}.csv'
    )
    cerebro.adddata(data, name=symbol)

# Access in strategy
class MyStrategy(bt.Strategy):
    def __init__(self):
        # Access by index
        self.data0 = self.datas[0]

        # Access by name
        self.aapl = self.getdatabyname('AAPL')