Table 2 Heterogeneity assessment using Bayesian model averaging

The table provides the results of meta-regressions using Bayesian model averaging. The dependent variable for each of the meta-regression models is the percentage change in emissions. The posterior inclusion probability (PIP) indicates the relevance of each variable. Variables with PIP≥0.5 are considered relevant for explaining the heterogeneity in carbon emissions reductions reported across primary studies. Post Mean and Post SD represent the mean and standard deviation of the posterior distribution for a respective explanatory variable. Five variables have PIP≥0.5 and are considered relevant (marked in bold): the dummy variables for RGGI, Chinese pilot ETS, Swiss ETS, Data_City, and duration. The dummy variables represent the geographic location in which the policy was implemented, with the reference location being EU ETS. Data_City captures whether primary studies used city level data versus country level data. The variable duration captures the number of years for which data on the scheme was collected after the policy was implemented. Definitions of the other explanatory variables are provided in the Supplementary Information.

Search