Luke Taylor - Aarhus University
"Type I and Type II Error Probabilities in the Courtroom"
Abstract
We estimate the probability of a miscarriage of justice using a novel nonparametric estimator for misclassified binary choice models. We show how to test the required large support condition, investigate the validity of the exclusion restriction, and give an alternative estimator for when the large support test fails. Depending on the defendant’s race and gender, the probability of wrongful conviction is estimated to range from 63% to 76%, and wrongful acquittal from 4% to 6.5%. These surprising results are due to our sample only including defendants who have reached a final trial, implying substantial evidence against them.
Additional information:
- Speaker: Luke Taylor
- Time: Thursday, 30.11.2023, 15:00 - 16:00
- Location: Faculty Lounge, Room 0.036 / Online via Zoom
- Further links:
- Organizer: Statistics Group
- Contact:
- Almut Lunkenheimer
- +49 228 73-9228
- ifs@uni-bonn.de