Reading some of the online links related to interview questions for data scientist position , i stumbled upon a question – what is the difference between Machine Learning and Econometrics? Coming from economics background one always feels that there is a lot of similarity between the two but the discussion below provides a good overview
Varian(2014), Diebold(2016) and Mullainathan & Spiess (2017) have written blog posts and articles which discuss this question in detail. Following posts discusses a general theme and readers curious to learn more should definitely read the articles referenced here.
- Machine learning is mostly concerned with both summarizing data and prediction (i.e ) whereas Econometrics can be broken down into four main categories prediction, summarization, estimation and inference.
- Machine learning is concerned with non-causal prediction whereas in econometrics we are ore concerned with causal prediction. Diebold(2016) in his posts explains machine learning non-causal prediction as “if a new person arrives with covariates what is my minimum – Mean Squared Error guess of ? ” whereas causal prediction is more concerned with, if i intervene and give a person a certain treatment what is my minimum- Mean Squared Error guess of ?
- Machine learning often uses the terms features or predictors which in econometrics is known as independent variables.
- 2014. “Big Data: New Tricks for Econometrics.” Journal of Economic Perspectives, 28 (2): 3-28. Can be downloaded here.
- Diebold : https://fxdiebold.blogspot.com/2016/10/machine-learning-vs-econometrics-i.html
- 2017. “Machine Learning: An Applied Econometric Approach.”Journal of Economic Perspectives, 31 (2): 87-106