How to Build a Winning Machine Learning FOREX Strategy in Python: Getting & Plotting Historical Data

In this video we are going learn how about the various sources for historical FOREX data. Primarily, we will be using data from Dukascopy bank. There are many other FOREX historical data sources that you need to pay for that will be of much higher quality. However, I feel that the data available for free at Dukascopy will be quite sufficient for our uses. Additionally, I give a demonstration on how to set up a plot that, in my opinion, is much easier/better looking than a plot using Matplotlib. Plotly is very flexible and can be used to generate many different plot types; I strongly recommend it. In the next video we will begin to construct functions that will return financial indicators. Down the road, we will use the indicators to train a machine learning algorithm to make binary price predictions. We will use a test dataset to backtest our strategy and adjust parameters to our liking. Useful links: **** DUKASCOPY HISTORICAL DATA FEED **** https://www.dukascopy.com/swiss/english/marketwatch/historical/ **** PYTHON FFN: FINANCIAL FUNCTIONS **** http://pmorissette.github.io/ffn/ **** PLOTLY **** https://plot.ly/ **** PANDAS DOCUMENTATION **** http://pandas.pydata.org/pandas-docs/stable/ **** TRADING VIEW **** https://www.tradingview.com/