py-metatrader

.. image:: https://badges.gitter.im/Join%20Chat.svg :alt: Join the chat at hhttps://gitter.im/samuraitaiga/py-metatrader :target: https://gitter.im/samuraitaiga/py-metatrader

py-metatrader 0.0.1

Released: 30-May-2015

============= Introduction

py-metatrader is a python package that provides interfaces to metatrader4(mt4). metatrader4_ is a trading platform that can automate trading(fx, stock, etc...) by your own program(ExpertAdvisor in mt4).

you can automate simuration(backtest in mt4), CI EA development , etc... by using this library.

currently works with Python 2.7.

contributing and porting is welcome.

============= Feature

At the moment, py-metatrader supports:

• backtest
• optimization

The goal of py-metatrader is to support execute all feature of metatrader4 from this library.

============ Installation

Install via pip_:

.. code-block:: bash

$ pip install metatrader

Install from source:

.. code-block:: bash

$ git clone https://github.com/samuraitaiga/py-metatrader.git
$ cd py-metatrader
$ python setup.py install

============ ChangeLogs

• 0.0.1

  • first release. backtest and optimization from python.

============ Usage

Backtest:

.. code-block:: python

from metatrader.mt4 import initizalize
from metatrader.backtest import BackTest

# point mt4 install folder
initizalize('C:\\Program Files\\FXCM MetaTrader 4')

# specify backtest period by datetime format
from_date = datetime(2014, 9, 1)
to_date = datetime(2015, 1, 1)

ea_name = 'Moving Average'

# create ea param by dict.
param = {
         'Lots': {'value': 0.1},
         'MaximumRisk': {'value': 0.02},
         'DecreaseFactor': {'value': 3.0},
         'MovingPeriod': {'value': 12},
         'MovingShift': {'value': 6}
         }
# create backtest object
backtest = BackTest(ea_name, param, 'USDJPY', 'M5', from_date, to_date)

# create backtest object with specified spread
# backtest = BackTest(ea_name, param, 'USDJPY', 'M5', from_date, to_date, spread=10)

# run backtest
ret = backtest.run()

# you can get result from result object
# for example you can print gross profit
print ret.gross_profit

.. _metatrader4: http://www.metatrader4.com/ .. _pip: http://www.pip-installer.org/