Metodo

International Studies in Phenomenology and Philosophy

Book | Chapter

196139

Can graphical causal inference be extended to nonlinear settings?

Nadine ChlaßAlessio Moneta

pp. 63-72

Abstract

This paper assesses an extension of the method for graphical causal inference proposed by Spirtes et al. and Pearl to nonlinear settings. We propose nonparametric tests for conditional independence based on kernel density estimation and study their relative performance in a Monte Carlo study. Our method outperforms Fischer's z test for nonlinear settings while subject to the so-called curse of dimensionality. We do show, however, how the latter can be overcome by using local bootstrapping.

Publication details

Published in:

Surez Mauricio, Dorato Mauro, Rédei Miklós (2010) Epsa epistemology and methodology of science: launch of the European philosophy of science association. Dordrecht, Springer.

Pages: 63-72

DOI: 10.1007/978-90-481-3263-8_6

Full citation:

Chlaß Nadine, Moneta Alessio (2010) „Can graphical causal inference be extended to nonlinear settings?“, In: M. Surez, M. Dorato & M. Rédei (eds.), Epsa epistemology and methodology of science, Dordrecht, Springer, 63–72.