Clustering methods are often used to model energy consumption for two reasons. First, clustering is often used to process data and to improve the predictive accuracy of subsequent energy models. Second, stable clusters that are reproducible with respect to non-essential changes can be used to group, target, and interpret observed subjects. However, it is well known that clustering methods are highly sensitive to the choice of algorithms and variables. This can lead to misleading assessments of predictive accuracy and mis-interpretation of clusters in policymaking.This paper therefore introduces two methods to the modeling of energy consumption in buildings: clusterwise regression, also known as latent class regression, which integrates clustering and regression simultaneously; and cluster validation methods to measure stability. Using a large dataset of multifamily buildings in New York City, clusterwise regression is compared to common two-stage algorithms that use K-means and model-based clustering with linear regression. Predictive accuracy is evaluated using 20-fold cross validation, and the stability of the perturbed clusters is measured using the Jaccard coefficient. These results show that there seems to be an inherent tradeoff between prediction accuracy and cluster stability. This paper concludes by discussing which clustering methods may be appropriate for different analytical purposes

Many U.S. states have taken significant action on climate change in recent years, demonstrating their commitment despite federal policy gridlock and rollbacks. Yet, there is still much we do not know about the agents, discourses, and strategies of those seeking to delay or obstruct state-level climate action. We first ask, what are the obstacles to strong and effective climate policy within U.S. states?We review the political structures and interest groups that slow action, and we examine emerging tensions between climate justice and the technocratic and/or market-oriented approaches traditionally taken by many mainstream environmental groups. Second,what are potential solutions for overcoming these obstacles?We suggest strategies for overcoming opposition to climate action that may advance more effective and inclusive state policy, focusing on political strategies, media framing, collaboration, and leveraging the efforts of ambitious local governments.

Many different governments have begun to require disclosure of building energy performance, in order to allow owners and prospective buyers to incorporate this information into their investment decisions. These policies, known as disclosure or information policies, require owners to benchmark their buildings and sometimes conduct engineering audits. However, given substantial variation in the cost to disclose different types of information, it is natural to ask: how much and what kind of information about building energy performance should be disclosed, and for what purposes? To answer this question, this paper assembles and cleans a comprehensive panel dataset of New York City multifamily buildings, and analyzes its predictive power using a Bayesian multilevel regression model. Analysis of variance (ANOVA) reveals that building-level variation is the most important factor in explaining building energy use, and that there are few, if any, relationships of building systems to observed energy use. This indicates that disclosure laws requiring benchmarking data may be relatively more useful than engineering audits in explaining the observed energy performance of existing buildings. These results should inform the further development of information disclosure laws

Statistical models can only be as good as the data put into them. Data about energy consumption continues to grow, particularly its non-technical aspects, but these variables are often interpreted differently among disciplines, datasets, and contexts. Selecting key variables and interactions is therefore an important step in achieving more accurate predictions, better interpretation, and identification of key subgroups for further analysis.This paper therefore makes two main contributions to the modeling and analysis of energy consumption of buildings. First, it introduces regularization, also known as penalized regression, for principled selection of variables and interactions. Second, this approach is demonstrated by application to a comprehensive dataset of energy consumption for commercial office and multifamily buildings in New York City. Using cross-validation, this paper finds that a newly-developed method, hierarchical group-lasso regularization, significantly outperforms ridge, lasso, elastic net and ordinary least squares approaches in terms of prediction accuracy; develops a parsimonious model for large New York City buildings; and identifies several interactions between technical and non-technical parameters for further analysis, policy development and targeting. This method is generalizable to other local contexts, and is likely to be useful for the modeling of other sectors of energy consumption as well