Machine learning and structural econometrics: contrasts and synergies

Abstract

Summary We contrast machine learning (ML) and structural econometrics (SE), focusing on areas where ML can advance the goals of SE. Our views have been informed and inspired by the contributions to this special issue and by papers presented at the second conference on dynamic structural econometrics at the University of Copenhagen in 2018, ‘Methodology and Applications of Structural Dynamic Models and Machine Learning'. ML offers a promising class of techniques that can significantly extend the set of questions we can analyse in SE. The scope, relevance and impact of empirical work in SE can be improved by following the lead of ML in questioning and relaxing the assumption of unbounded rationality. For the foreseeable future, however, ML is unlikely to replace the essential role of human creativity and knowledge in model building and inference, particularly with respect to the key goal of SE, counterfactual prediction.

Does climate help modeling COVID-19 risk and to what extent?

“…Of particular interest is the complementarity between machine learning and econometric approaches, where the first is intent on prediction while the latter focuses on explanation. As discussed in the literature, the use of a hybrid approach where machine learning modeling is paired with econometric analysis can help address relative weaknesses in the two methods by leveraging relative strengths [69][70][71][72][73]. For example, machine learning is better equipped to take advantage of structural heterogeneity in training data to make short-term predictions, whereas econometric methods are better at capturing long-term trends [73].…”

Structural Models: Inception and Frontier

“…We also believe that there is much scope for structural models in many applied microeconomic fields to integrate with newly popular machine learning techniques in the same way consumer choice models in marketing and industrial organization and causal inference approaches are already integrating. Iskhakov et al (2020) provide a survey of this new literature that explores possible synergies between structural econometrics and machine learning.…”

Hardness of Learning in Rich Environments and Some Consequences for Financial Markets

“…This paper is part of a recent literature that extends tools from theoretical machine learning to elucidate issues in econometrics and economic learning theory (Athey and Wagner [4], Chernozhukov et al [5], Dao et al [6], Iskhakov et al [7], Oprescu et al [8], Singh et al [9], Sverdrup et al [10], Syrgkanis et al [11], Syrgkanis and Zampetakis [12]).…”