==================
- N.B.: This version supercedes the prior pre-print of the paper, which can be found at mjbommar/scotus-predict.
- Title: A General Approach for Predicting the Behavior of the Supreme Court of the United States
- Authors: Daniel Martin Katz, Michael J Bommarito II, Josh Blackman
- Paper URL: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2463244
- Blog URL: http://lexpredict.com/portfolio/predicting-the-supreme-court/
Building on developments in machine learning and prior work in the science of judicial prediction, we construct a model designed to predict the behavior of the Supreme Court of the United States in a generalized, out-of-sample context. Our model leverages the random forest method together with unique feature engineering to predict nearly two centuries of historical decisions (1816-2014). Using only data available prior to decision, our model outperforms null (baseline) models at both the justice and case level under both parametric and non-parametric tests. Over nearly two centuries, we achieve 70.2% accuracy at the case outcome level and 71.9% at the justice vote level. More recently, over the past century, we outperform a high quality null model by nearly 5%. Our performance is consistent with, but improves upon, the general level of prediction demonstrated by prior work; however, our model is distinctive because it can be applied out-of-sample to the entire past and future of the Court, not just one year. Our results represent an advance for the science of quantitative legal prediction and portend a range of other potential applications.
The source and data in this repository allow for the reproduction of the results in this paper.
- Model run used for publication figures: https://github.com/mjbommar/scotus-predict-v2/blob/master/src/model_growing_random_forest_cv_5.ipynb
- "Always guess reverse" model: https://github.com/mjbommar/scotus-predict-v2/blob/master/src/baseline_model_always_reverse.ipynb
- Sample alternative model run: https://github.com/mjbommar/scotus-predict-v2/blob/master/src/model_growing_random_forest_1.ipynb
- Disposition coding map: https://github.com/mjbommar/scotus-predict-v2/blob/master/src/legacy_model.py#L73
- Publication figures: https://github.com/mjbommar/scotus-predict-v2/blob/master/src/publication_figures.ipynb
The data used in this paper is available from the Supreme Court Database (SCDB); both the Modern and Legacy databases were used in this analysis.
The latest version of this model was released in December 2016.