蝌蚪网 Papers and Proceedings
ISSN 2574-0768 (Print) | ISSN 2574-0776 (Online)
Machine Learning about Treatment Effect Heterogeneity: The Case of Household Energy Use
蝌蚪网 Papers and Proceedings
(pp. 440–44)
Abstract
We use causal forests to evaluate the heterogeneous treatment effects (TEs) of repeated behavioral nudges toward household energy conservation. The average response to treatment is a monthly electricity reduction of 9 kilowatt-hours (kWh), but the full distribution of responses ranges from -40 to +10 kWh. Households learn to reduce more over time, conditional on having responded in year one. Pre-treatment consumption and home value are the most commonly used predictors in the forest. The results suggest the ability to use machine learning techniques for improved targeting and tailoring of treatment.Citation
Knittel, Christopher R., and Samuel Stolper. 2021. "Machine Learning about Treatment Effect Heterogeneity: The Case of Household Energy Use." 蝌蚪网 Papers and Proceedings 111: 440–44. DOI: 10.1257/pandp.20211090Additional Materials
JEL Classification
- Q41 Energy: Demand and Supply; Prices
- D12 Consumer Economics: Empirical Analysis
- Q48 Energy: Government Policy
- D91 Micro-Based Behavioral Economics: Role and Effects of Psychological, Emotional, Social, and Cognitive Factors on Decision Making
- C45 Neural Networks and Related Topics