Multicollinearity in the Presence of Errors-in-Variables Can Increase the Probability of Type-I Error

16 Pages Posted: 21 Feb 2019

See all articles by John Komlos

John Komlos

Ludwig Maximilian University of Munich (LMU) - Faculty of Economics; CESifo (Center for Economic Studies and Ifo Institute)

Date Written: 2019

Abstract

Multicollinearity, especially in combination with errors-in-variables, can increase the likelihood of a Type-I error by inflating the value of the estimated coefficients by more than it magnifies their standard errors, thereby increasing the likelihood of obtaining statistically significant results. This anomalous result may be due to an interaction effect between errors-in-variables and multicollinearity.

Keywords: multicollinearity, Type I error, errors-in-variables

JEL Classification: C010

Suggested Citation

Komlos, John, Multicollinearity in the Presence of Errors-in-Variables Can Increase the Probability of Type-I Error (2019). CESifo Working Paper No. 7459. Available at SSRN: https://ssrn.com/abstract=3338813

John Komlos (Contact Author)

Ludwig Maximilian University of Munich (LMU) - Faculty of Economics ( email )

Ludwigstrasse 28
Munich, D-80539
Germany

CESifo (Center for Economic Studies and Ifo Institute)

Poschinger Str. 5
Munich, DE-81679
Germany

Register to save articles to
your library

Register

Paper statistics

Downloads
19
Abstract Views
77
PlumX Metrics