Goodhart's law and machine learning: a structural perspective

A. Hennessy, Christopher and Goodhart, C. A. E. (2023) Goodhart's law and machine learning: a structural perspective. International Economic Review. ISSN 1468-2354

Text (Goodhart's law and machine learning. A structural perspective.) - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (116kB)

• Author

Abstract

We develop a simple structural model to illustrate how penalized regressions generate Goodhart bias when training data are clean but covariates are manipulated at known cost by future agents. With quadratic (extremely steep) manipulation costs, bias is proportional to Ridge (Lasso) penalization. If costs depend on absolute or percentage manipulation, the following algorithm yields manipulation-proof prediction: Within training data, evaluate candidate coefficients at their respective incentive-compatible manipulation configuration. We derive analytical coefficient adjustments: slopes (intercept) shift downward if costs depend on percentage (absolute) manipulation. Statisticians ignoring manipulation costs select socially suboptimal penalization. Model averaging reduces these manipulation costs.

Item Type:	Article
Additional Information:	© 2023 The Author(s).
Divisions:	Financial Markets Group
Subjects:	H Social Sciences > HB Economic Theory
JEL classification:	C - Mathematical and Quantitative Methods > C0 - General > C00 - General
Date Deposited:	18 Apr 2023 13:12
Last Modified:	17 Oct 2024 18:01
URI:	http://eprints.lse.ac.uk/id/eprint/118656

Actions (login required)

View Item