We propose the use of Kernel Regularized Least Squares (KRLS) for social science modeling and inference problems. KRLS borrows from machine learning methods designed to solve regression and classification problems without relying on linearity or additivity assumptions. The method constructs a flexible hypothesis space that uses kernels as radial basis functions and finds the best-fitting surface in this space by minimizing a complexity-penalized least squares problem. We argue that the method is well-suited for social science inquiry because it avoids strong parametric assumptions, yet allows interpretation in ways analogous to generalized linear models while also permitting more complex interpretation to examine non-linearities and heterogeneous effects. We also extend the method in several directions to make it more effective for social inquiry, by (1) deriving estimators for the pointwise marginal effects and their variances, (2) establishing unbiasedness, consistency, and asymptotic normality of the KRLS estimator under fairly general conditions, (3) proposing and justifying a simple automated rule for choosing the kernel bandwidth, and (4) providing companion software. We illustrate the use of the method through several simulations and a real-data example. Jens Hainmueller, Department of Political Science, 77 Massachusetts Avenue, Cambridge, MA 02139. E-mail: jhainm@mit.edu. Chad Hazlett, Department of Political Science, 77 Massachusetts Avenue, Cambridge, MA 02139. E-mail: hazlett@mit.edu. Authors are listed in alphabetical order and contributed equally. We thank Jeremy Ferwerda, Dominik Hangartner, Lorenzo Rosasco, Marc Ratkovic, Teppei Yamamoto, and participants in seminars at NYU, MIT, the Midwest Political Science Conference, and the European Political Science Association Conference for helpful comments. Companion software written by the authors to implement the methods proposed in this paper in R, Matlab, and Stata can be downloaded at http://www.mit.edu/ jhainm/software.htm. The usual disclaimer applies.
[1]
Tomaso A. Poggio,et al.
Regularization Networks and Support Vector Machines
,
2000,
Adv. Comput. Math..
[2]
A. Atiya,et al.
Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond
,
2005,
IEEE Transactions on Neural Networks.
[3]
Yong Wang,et al.
Model Selection for Regularized Least-Squares Classification
,
2005,
ICNC.
[4]
Gene H. Golub,et al.
Generalized cross-validation as a method for choosing a good ridge parameter
,
1979,
Milestones in Matrix Computation.
[5]
R. Friedrich.
In Defense of Multiplicative Terms In Multiple Regression Equations
,
1982
.
[6]
G. King,et al.
Improving Quantitative Studies of International Conflict: A Conjecture
,
2000,
American Political Science Review.
[7]
A Tikhonov,et al.
Solution of Incorrectly Formulated Problems and the Regularization Method
,
1963
.
[8]
Alexander Gammerman,et al.
Ridge Regression Learning Algorithm in Dual Variables
,
1998,
ICML.