Introduction
I believe that machine learning (ML) will have a dramatic impact on the field of economics within a short time frame. Indeed, the impact of ML on economics is already well underway, and so it is perhaps not too difficult to predict some of the effects.
This paper begins by stating the definition of ML that I will use in this paper, describing its strengths and weaknesses, and contrasting ML with traditional econometrics tools for causal inference, which is a primary focus of the empirical economics literature. Next, I review some applications of ML in economics where ML can be used off-the-shelf: the use case in economics is essentially the same use case that the ML tools were designed an optimized for. I then review "prediction theory" problems (Kleinberg et al., 2015), where prediction tools have been embedded in the context of economic decision-making. Then, I provide an overview of the questions considered and early themes of the emerging literature in econometrics and statistics combining machine learning and causal inference, a literature that is providing insights and theoretical results that are novel from the perspective of both ML and statistics/econometrics. Finally, I step back and describe the implications of the field of economics as a whole.
The paper highlights several themes.
A first theme is that ML does not add much to questions about identification, which concern when the object of interest, e.g., a causal effect, can be estimated with infinite data, but rather yields great improvements when the goal is semi-parametric estimation or when there are a large number of covariates relative to the number of observations. ML has great strengths in using data to select functional forms flexibly.
A second theme is that a key advantage of ML is that ML views empirical analysis as "algorithms" that estimate and compare many alternative models. This approach contrasts with economics, where (in principle, though rarely in reality) the researcher picks a model based on principles and estimates it once. Instead, ML algorithms build in "tuning" as part of the algorithm. The tuning is essentially model selection, and in an ML algorithm that is data-driven. There are a whole host of advantage of this approach, including improved performance as well as enabling researchers to be systematic and fully describe the process by which their model was selected. Of course, cross-validation has also been used historically in economics, for example for selecting the bandwidth for a kernel regression, but it is viewed as a fundamental part of an algorithm in ML.
(...)
A fourth theme is that the algorithms also have to be modified to provide valid confidence intervals for estimated effects when the data is used to select the model. Many recent papers make use of techniques such as sample splitting, leave-one-out estimation, and other similar techniques to provide confidence intervals that work both in theory and in practice. The upside is that using ML can provide the best of both worlds: the model selection is data driven, systematic, and a wide range of models are considered; yet, the model selection process is fully documented, and confidence intervals take into account the entire algorithm.
Finally, the combination of ML and newly available datasets will change economics in fairly fundamental ways, ranging from new questions, to new approaches to collaboration (larger teams and interdisciplinary interaction), to a change in how involved economists are in the engineering and implementation of policies.
2. What is Machine Learning and What are Early Use Cases?
It is harder than one might think to come up with an operational definition of ML. The term can be (and has been) used broadly or narrowly; it can refer to a collections of subfields of computer science, but also to a set of topics that are developed and used across computer science, engineering, statistics, and increasingly the social sciences. Indeed, one could devote an entire article to the definition of ML, or to the question of whether the thing called ML really needed a new name other than statistics, the distinction between ML and AL, and so on. However, I will leave this debate others, and focus on a narrow, practical definition that will make it easier to distinguish ML from the most commonly used econometric approaches used in applied econometrics until very recently. For readers coming from a machine learning background, it is also important to note that applied statistics and econometrics have developed a body of insights on topics ranging from causal inference to efficiency that have not yet been incorporated in mainstream machine learning, while other parts of machine learning have overlap with methods that have been used in applied statistics and social sciences for many decades.
Starting from a relatively narrow definition of machine learning, machine learning is a field that develops algorithms designed to be applied to datasets, with the main areas of focus being prediction (regression), classification, and clustering or grouping tasks. These tasks are divided into two main branches, supervised and unsupervised ML.
(...)
Supervised machine learning typically entails using a set of features or covariates (X) to predict an outcome (Y). When using the term prediction, it is important to emphasize that framework focuses not on forecasting, but rather on a setting where there are some labelled observations where both X and Y are observed (the training data), and the goal is to predict outcomes (Y) in an independent test set based on the realized values of X for each unit in the test set. In other words, the goal is to construct mu hat (x), which is an estimator of mu (x) = E[Y|X = x], in order to do a good job predicting the true values
Comments
Post a Comment