Problems with Statistical Tests of Atmospheric Greenhouse Gas Concentration Data Derived from Ice Cores

38 Pages Posted: 5 Jan 2019 Last revised: 8 Feb 2022

See all articles by James Ross McCown

James Ross McCown

University of Oklahoma - Division of Finance; Toltec Group

Multiple version iconThere are 2 versions of this paper

Date Written: February 6, 2022

Abstract

Direct measurements of the concentrations of greenhouse gases were only made sporadically before 1958 for carbon dioxide, 1978 for nitrous oxide, and 1983 for methane. Prior to those dates, several researchers used concentrations derived from ice cores that have poor temporal accuracy. Several of the studies concluded that radiative forcing from the greenhouse gases is cointegrated with atmospheric temperatures. Monte Carlo simulations of statistical tests of data derived from ice cores are shown to yield spurious results when conducted using annual data. Claims of cointegration between radiative forcing of greenhouse gases and temperature are likely invalid.

Keywords: Greenhouse gases, carbon dioxide, unit roots, cointegration

JEL Classification: Q54, C22

Suggested Citation

McCown, James Ross, Problems with Statistical Tests of Atmospheric Greenhouse Gas Concentration Data Derived from Ice Cores (February 6, 2022). Available at SSRN: https://ssrn.com/abstract=3304901 or http://dx.doi.org/10.2139/ssrn.3304901

James Ross McCown (Contact Author)

University of Oklahoma - Division of Finance ( email )

Norman, OK 73019
United States

Toltec Group ( email )

Oklahoma City, OK
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
189
Abstract Views
1,369
Rank
285,122
PlumX Metrics