Big Data and Application of Machine Learning in Economics

Catalog description

This econometrics course introduces the concept of big data and econometric techniques to analyze big data using the tools of machine learning, and provides main ideas and insights on how we can use big data to solve economic problems. The course discusses differences in objectives, techniques, and settings between the machine learning literature in computer science and economics, and introduces some specific methods from the machine learning literature that are emerging as important tools for economists. These methods include supervised learning for regression and classification, ridge regression estimator, the lasso regression estimator, random forest, dimensionality reduction, unsupervised learning methods of clustering, and natural language processing and data scraping as part of data collection. The course introduces students to these methods primarily using Stata and the integration of Stata with Python. The course does not require prior knowledge in computer programming, but it requires standard knowledge in econometrics.