SOC-PHSYSYOct 25, 2018

Estimating Heterogeneous Treatment Effects in Residential Demand Response

arXiv:1710.031901 citationsh-index: 84
AI Analysis

For utility companies and policymakers, this work provides a method to identify households that respond best to price interventions, potentially improving cost efficiency of demand response programs.

This paper estimates the causal effect of hour-ahead price interventions on residential electricity consumption using data from 7,000 California households, finding an average treatment effect of ~0.10 kWh (11%) per intervention. It uses causal decision trees to detect treatment effect heterogeneity and shows their superiority over k-means clustering for targeting households.

We evaluate the causal effect of hour-ahead price interventions on the reduction in residential electricity consumption using a data set from a large-scale experiment on 7,000 households in California. By estimating user-level counterfactuals using time-series prediction, we estimate an average treatment effect of ~0.10 kWh (11%) per intervention and household. Next, we leverage causal decision trees to detect treatment effect heterogeneity across users by incorporating census data. These decision trees depart from classification and regression trees, as we intend to estimate a causal effect between treated and control units rather than perform outcome regression. We compare the performance of causal decision trees with a simpler, yet more inaccurate k-means clustering approach that naively detects heterogeneity in the feature space, confirming the superiority of causal decision trees. Lastly, we comment on how our methods to detect heterogeneity can be used for targeting households to improve cost efficiency.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes