A new research paper from the IMF outlines some of the implications for Machine Learning in casual financial analysis models and the impact of the technology in the case of a hypothetical financial crisis.
"Machine learning tools are well known for their success in prediction. But prediction is not causation, and causal discovery is at the core of most questions concerning economic policy," Andrew J Tiffin, the report's author notes.
Examining leading research in the area of Machine Learning, Tiffin notes that literature has typically focused more on issues of causality and instead introduces some leading work in this area, using a concrete example—assessing the impact of a hypothetical banking crisis on a country’s growth.
Machine-learning literature has typically concerned itself mostly with prediction, but has recently focused more and more on issues of causality—prompted in part by a growing realization that many problems originally thought to be predictive (e.g., what happens to online sales if we increase a particular form of advertising) actually entail the discovery of a more causal relationship.
"By enabling consideration of a rich set of potential nonlinearities, and by allowing individually-tailored policy assessments, machine learning can provide an invaluable complement to the skill set of economists within the Fund and beyond" Tiffin concludes.
Read this working paper in full: http://www.imf.org/external/pubs/cat/longres.aspx?sk=48722
IMF Working Papers describe research in progress by the author(s) and are published to elicit comments and to encourage debate. The views expressed in IMF Working Papers are those of the author(s) and do not necessarily represent the views of the IMF, its Executive Board, or IMF management. For a full list of IMF Publications visit: https://www.imf.org/en/Publications