- The goal is to provide a backtesting and live trading tool, which can run multiple strategies on up to 10 currency pairs.
- Strategies can monitor up to three timeframes (e.g. H1, H2 and D1) and calculate buy/sell actions based on them.
Besides building your own, you can also use a wide range of indicators via TA-Lib or numpy.
- OANDA supports most commonly used timeframes. You can find all supported values here.
- Add Matplot/Plot.ly support. See also following tutorial.
- Implement proxy class for backtesting, which will first check if data is available locally and only then fetch/save via API.
- Implement usage of ETags to reduce traffic/latency.
- Prepare Makefile.
sudo apt-get install virtualenv python-setuptools libssl-dev:
sudo apt-get build-dep python-matplotlib
Additionally get TA-Lib from http://www.ta-lib.org/hdr_dw.html
git clone git@github.com:jmelett/pyfx.git cd pyfx virtualenv env source env/bin/activate pip install -e . envdir .my_envs python ./_cmd.py -h
In case you encounter the error `ImportError: libta_lib.so.0: cannot open shared object file: No such file or directory`
Following command* will do the trick: `export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH`
* See also http://stackoverflow.com/questions/11813279/python-wrapper-for-ta-lib-import-failure