Recidivism Prediction, Peer Effect Estimation, and Prediction-Powered Inference with LLM Text Measures
A novel instrumental variable estimator is developed that accommodates multivariate outcomes, sparse networks, and multidimensional latent homophily and is shown to be $\sqrt{N}$-consistent and asymptotically normal under sparsity conditions that relax dense-network assumptions prevalent in the peer effect literature.
Shanjukta Nath, Jiwon Hong, Jae Ho Chang et al.
· 0 citations