Department of Economics, and Institute for New Economic Thinking at the Oxford Martin School, University of Oxford, Oxford, UK
Received: 16 Jan 2013 – Discussion started: 06 Feb 2013
Abstract. We outline six important hazards that can be encountered in econometric modelling of time-series data, and apply that analysis to demonstrate errors in the empirical modelling of climate data in Beenstock et al. (2012). We show that the claim made in Beenstock et al. (2012) as to the different degrees of integrability of CO2 and temperature is incorrect. In particular, the level of integration is not constant and not intrinsic to the process. Further, we illustrate that the measure of anthropogenic forcing in Beenstock et al. (2012), a constructed "anthropogenic anomaly", is not appropriate regardless of the time-series properties of the data.
Revised: 28 Aug 2013 – Accepted: 09 Sep 2013 – Published: 22 Oct 2013
Pretis, F. and Hendry, D. F.: Comment on "Polynomial cointegration tests of anthropogenic impact on global warming" by Beenstock et al. (2012) – some hazards in econometric modelling of climate change, Earth Syst. Dynam., 4, 375-384, doi:10.5194/esd-4-375-2013, 2013.