Generalized Kernel Regularized Least Squares
Political Analysis · 2024-04 · 32 (2) · 157-171
DOI: 10.1017/pan.2023.27 ↗Kurz gefasst
Fachartikel in der Zeitschrift Political Analysis, online erschienen am 01.09.2023. Verfasst von Qing Chang und 1 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Kernel regularized least squares (KRLS) is a popular method for flexibly estimating models that may have complex relationships between variables. However, its usefulness to many researchers is limited for two reasons. First, existing approaches are inflexible and do not allow KRLS to be combined with theoretically motivated extensions such as random effects, unregularized fixed effects, or non-Gaussian outcomes. Second, estimation is extremely computationally intensive for even modestly sized datasets. Our paper addresses both concerns by introducing generalized KRLS ( gKRLS ). We note that KRLS can be re-formulated as a hierarchical model thereby allowing easy inference and modular model construction where KRLS can be used alongside random effects, splines, and unregularized fixed effects. Computationally, we also implement random sketching to dramatically accelerate estimation while incurring a limited penalty in estimation quality. We demonstrate that gKRLS can be fit on datasets with tens of thousands of observations in under 1 min. Further, state-of-the-art techniques that require fitting the model over a dozen times (e.g., meta-learners) can be estimated quickly.
Zitieren
APA
Chang, Q., & Goplerud, M. (2024). Generalized Kernel Regularized Least Squares. Political Analysis, 32(2), 157-171. https://doi.org/10.1017/pan.2023.27
Vancouver
Chang Q, Goplerud M. Generalized Kernel Regularized Least Squares. Political Analysis. 2024;32(2):157-171. doi:10.1017/pan.2023.27
BibTeX
@article{Chang2024,
author = {Chang, Qing and Goplerud, Max},
title = {Generalized Kernel Regularized Least Squares},
journal = {Political Analysis},
year = {2024},
volume = {32},
number = {2},
pages = {157--171},
doi = {10.1017/pan.2023.27}
}
EUVIMED