Machine learning techniques are increasingly being evaluated in academia while also being leveraged by practitioners at policy institutions such as central banks and governments. A themed issue of the Journal of Econometrics aims to present frontier research on machine learning and economic policy.
The journal has invited paper submissions on ‘Machine learning for economic policy,’ to be submitted by May 31, 2023.
The themed issue will cover a wide range of applications and methodological contributions, including deep learning, text analytics, reinforcement learning, shock identification, forecasting and nowcasting, identification, and various approaches to model interpretability and inference.
There are good reasons for policymakers to embrace these new techniques. Tree-based models or artificial neural networks, when combined with novel and rich data sources such as text or high-frequency indicators, can provide prediction accuracy and information that traditional models cannot. For example, machine learning can reveal previously unknown but critical nonlinearities in the data generation process. Moreover, natural language processing—made possible by advances in machine learning—is increasingly being applied to better understand the economic landscape that policymakers must survey.
The upsides of these new techniques come with the downside that it often is not clear what is the mechanism through which the machine learning model operates. Much of the existence of the black box critique is due to how machine learning models evolved with a focus on accuracy. However, this single focus can be particularly problematic in decision-making situations, where all stakeholders have an interest in understanding all pieces of information which enter the decision-making process, irrespective of model accuracy. The tools of economics and econometrics can help to address this problem thereby building bridges between disciplines.
