Spurious Regressions and Near-Multicollinearity, with an Application to Aid, Policies and Growth

23 Pages Posted: 11 Nov 2012

See all articles by Jean-Bernard Chatelain

Jean-Bernard Chatelain

Paris School of Economics, Université Paris 1 Panthéon Sorbonne

Kirsten Ralf

Ecole Supérieure du Commerce Extérieur (ESCE)

Date Written: November 8, 2012

Abstract

In multiple regressions, explanatory variables with simple correlation coefficients with the dependent variable below 0.1 in absolute value (such as aid with economic growth) may have very large and statistically significant estimated parameters which are unfortunately "outliers driven" and spurious. This is obtained by including another regressor which is highly correlated with the initial regressor, such as a lag, a square or interaction terms of this regressor. The analysis is applied on the "Botswana outliers driven" Burnside and Dollar [2000] article which found that aid had an effect on growth only for countries achieving "good" macroeconomic policies.

Keywords: Near-Multicollinearity, Student t-Statistic, Spurious regressions, Ceteris paribus, Parameter Inflation Factor, Growth, Foreign Aid

JEL Classification: C12, C52, F35, O19, O47, P45

Suggested Citation

Chatelain, Jean-Bernard and Ralf, Kirsten, Spurious Regressions and Near-Multicollinearity, with an Application to Aid, Policies and Growth (November 8, 2012). Available at SSRN: https://ssrn.com/abstract=2173720 or http://dx.doi.org/10.2139/ssrn.2173720

Jean-Bernard Chatelain (Contact Author)

Paris School of Economics, Université Paris 1 Panthéon Sorbonne ( email )

48 Boulevard Jourdan
Paris, 75014 75014
France

HOME PAGE: http://www.parisschoolofeconomics.eu/fr/chatelain-jean-bernard/

Kirsten Ralf

Ecole Supérieure du Commerce Extérieur (ESCE) ( email )

10 Rue Sextius Michel
Paris, 75015
France

Register to save articles to
your library

Register

Paper statistics

Downloads
37
Abstract Views
695
PlumX Metrics