Installation
About
A novel bias-bound approach for non-parametric inference is introduced, focusing on both density and conditional expectation estimation. It constructs valid confidence intervals that account for the presence of a non-negligible bias and thus make it possible to perform inference with optimal mean squared error minimizing bandwidths. This package is based on Schennach (2020) doi:10.1093/restud/rdz065.
doi.org/10.1093/restud/rdz065 |
Key Metrics
Downloads
Yesterday | 4 0% |
Last 7 days | 17 -35% |
Last 30 days | 104 -30% |
Last 90 days | 387 -9% |
Last 365 days | 1.227 |
Maintainer
Maintainer | Xinyu DAI |