Decomposition analysis applied to energy: Some methodological issues

Abstract

Decomposition analysis is a systems analytical tool that has been widely applied in energy and environmental studies during the past two decades. The objective of this thesis is to study some methodological issues concerning this tool which have not been resolved in the past. These issues include consistency in aggregation, zero-negative values in the data set, perfect decomposition and integration of index decomposition analysis (IDA) and structure decomposition analysis (SDA). The log-mean Divisia method I (LMDI I) is proposed and proved to be perfect in decomposition that leaves no residue in the result. It is consistent in aggregation which allows for estimates for sub-groups to be aggregated in a consistent manner to a higher aggregation level. The modified Fisher ideal index (MFII) method is proposed to overcome the zero-negative value problem in the data set in both the multiplicative and additive forms. It is perfect in decomposition and robust to zero-negative values. IDA and SDA are compared in terms of the basic principle, historical development and applications. An integrated model (IM) that incorporates the desirable properties of both techniques is proposed.