Machine Learning and Trading Assets

Overview


Most of the techniques used to analyze market behaviour are drawn from standard statistics. In particular, OLS regression has become a workhorse of the financial quant community.

However, machine learning has made inroads into trading, in particular, in recognizing the trade patterns that are not easily captured by traditional techniques.

Deep Learning


Deep Learning is the most prominent of the machine learning techniques used in formulating trading decisions. They can be used to capture non-linear patterns in data that are not easily formalated as a standard regression problem.

The steps to building a deep learning trading algorithm are typically a variation of the following

  1. Feature Extraction
  2. Choosing the Network Type
  3. Choosing the Outputs
  4. Choosing the Layers