Comparing Old and New Partial Derivative Estimates from Nonlinear Nonparametric Regressions

14 Pages Posted: 15 Oct 2020 Last revised: 25 Jan 2021

See all articles by Hrishikesh D. Vinod

Hrishikesh D. Vinod

Fordham University - Department of Economics

Fred Viole

OVVO Financial Systems; Fordham University

Date Written: August 26, 2020

Abstract

Partial derivatives have a special place in economics since the marginal revolution of the 1850s. We present results from multivariate partial derivative estimates using nonlinear non-parametric regressions in a finite difference method, accessible via the R-package NNS. Numerical partial derivatives are notoriously unstable, but NNS always correctly estimates their sign and comes closest to the correct magnitude compared to the coefficients in multiple linear regressions, and compared to the gradients from the popular np package for non-parametric kernel regressions.

Keywords: multivariate regression, partial derivatives, finite difference

JEL Classification: C14, C30, C54, C60

Suggested Citation

Vinod, Hrishikesh D. and Viole, Fred, Comparing Old and New Partial Derivative Estimates from Nonlinear Nonparametric Regressions (August 26, 2020). Available at SSRN: https://ssrn.com/abstract=3681104 or http://dx.doi.org/10.2139/ssrn.3681104

Hrishikesh D. Vinod

Fordham University - Department of Economics ( email )

Dealy Hall
Bronx, NY 10458
United States
718-817-4065 (Phone)
718-817-3518 (Fax)

Fred Viole (Contact Author)

OVVO Financial Systems ( email )

NJ
United States

Fordham University ( email )

Bronx, NY 10458
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
43
Abstract Views
384
PlumX Metrics