Evaluation of an energy efficiency program in a regional context
- Authors: Martin, Peter , Lynch, David , Ali-Alkadiri, Mohammad , Lowe, Julian
- Date: 2011
- Type: Text , Conference paper
- Relation: 2011 International Energy Program Evaluation Conference: Impact through evaluation Boston, Massachusetts 16th-18th August, 2011
- Full Text:
- Reviewed:
- Description: The Central Victoria Solar City (CVSC) research trial is part of the Australian Government’s $94 million Solar Cities program. Managed by renewable energy company, Sustainable Regional Australia (SRA), the program encourages residents to test energy efficiency technologies and services designed to reduce energy use and reliance on non-renewable energy sources. The trial involves collecting data from over 3,500 households (including a control group of 750) across central Victoria and recording changes to their energy consumption until 30 June 2013. CVSC is in its early stages of implementation, with about one third of participants recruited. In energy program evaluations,much of the data is hierarchical in nature (e.g. household energy readings over time). An issue with such data is that conventional statistical methods (e.g. OLS or Logistic regression) assumeindependency between observations, which is likely to be violated by longitudinal data. Techniques to address this problem have been a major area of research during the past 10 years. Such developments have led to analytical tools (e.g. Linear Mixed Models), which allow for modeling of dependencies between measures. Early analysis has confirmed the hierarchical nature of the data, with 76% of the variance in pre-program energy consumption occurring between (rather than within) households. A preliminary baseline model based on regional, climatic and household characteristics explains 40% of variation in household electricity consumption. Initial findings suggest that household electricity consumption is most strongly influenced by regional factors (e.g. climate, reticulated gas availability), number of occupants, house size and income.
- Description: 2003009218
Marginal longitudinal semiparametric regression via penalized splines
- Authors: Ali-Alkadiri, Mohammad , Carroll, R.J. , Wand, M.P.
- Date: 2011
- Type: Text , Journal article
- Relation: Statistics and Probablitity Letters Vol. 80, no. 15-16 (2011), p. 1242-1252
- Full Text: false
- Reviewed:
- Description: We study the marginal longitudinal nonparametric regression problem and some of its semiparametric extensions. We point out that, while several elaborate proposals for efficient estimation have been proposed, a relative simple and straightforward one, based on penalized splines, has not. After describing our approach, we then explain how Gibbs sampling and the BUGS software can be used to achieve quick and effective implementation. Illustrations are provided for nonparametric regression and additive models.