Project-Pooled Analyses: Feasibility of Combining Data from Several Electric Utility Rate Demonstration Projects

摘要

The US-DOE has sponsored projects in numerous locations from Puerto Rico or Rhode Island to California on the effects of time-of-use pricing on electric power use by residential customers. Each project has analyzed its own data with inevitable variations in analysis techniques and interpretation between different projects. The feasibility of either separate but standardized analyses of each project's results and of pooling all the separate project results in order to perform one overall analysis is discussed. It is concluded that the rate structures and experimental designs of the demonstration projects are not sufficiently compatible to warrant pooled analyses. The analyst desiring to use a pooled data base is forced to construct artificial models that can be estimated. Numerous, untestable assumptions are required and little or no gain in overall precision can be expected. It is, therefore, recommended that the project-specific analyses be emphasized and that the future reporting requirements on the project-pooled analyses be eliminated. (ERA citation 06:003775)