Creative Commons

This dissertation was written as a part of the MSc in Energy Systems at the International Hellenic University. Climate change is a growing problem, studied extensively during the past few decades, focusing especially on our tampering with the environment. This dissertation attempts to augment the work of Stern and Kaufmann (2014) through several variables’ effects on temperature, with the implementation of more recent econometric techniques. Additionally, it concentrates on possible explanations behind dissimilar, with the aforementioned study, results, as well as on the outcomes’ progression over time. The analysis includes stationarity, cointegration and causality investigation, achieved with more than two tests in each case, between several gases radiative forcings and both HADCRUT4 and GISSv3 temperature time series, through a direct, in both completely and partially aggregated models, as well as through an indirect approach, in an entirely disaggregated model. Four scenarios are tested, and samples lie within the 1850 to 2011 and 1958 to 2011 time span. The investigation of the evolution of all the aforementioned causal relationships in all models and scenarios with time, using the fixed window on a rolling basis method, is considered a novelty as regards the climate change research. Results suggest that total, natural, anthropogenic and Greenhouse Gases’ radiative forcings cause temperature to change, while human induced sulfur emissions, solar irradiance and black carbon do not, throughout the largest part of the time period. Most of this research outcome is consistent with theory and Stern and Kaufmann (2014), with possible minor declination reasons the slightly different approach to Toda Yamamoto causality testing.