This record has been deprecated and the content removed

Read full text
Authors: Vansh Ruhela
Cited by: 0
References: 1

Abstract

A multi-domain scientific-computing toolkit for observational inference and intervention analysis, hosting the MRM (Multilevel Reconciliation Methodology) framework for Canadian carceral, police, and oversight data as its primary application. Provides general-purpose causal estimators (ATE, ATT, ATC, GATE, CATE, LATE, AIPW, G-computation), survey sampling methods (stratified, cluster, PPS, bootstrap, calibration), propensity-score and doubly-robust estimators, sensitivity analyses (E-value, Rosenbaum bounds), signal processing, statistical physics of crime (Hawkes self-exciting processes, reaction-diffusion, Lévy flight, urban scaling), and dataset ingestors for OTIS, SIU-IAP, Ontario SIU, TPS, CCRSO, plus US datasets (Chicago crime, NYC SQF, NIBRS, NamUs, NIST RDS) and Google BigQuery.

Fields of Study & Subjects

Advanced Causal Inference Techniques Scientific Computing and Data Management Optimal Experimental Design Methods

Publication details

Publication date
23 February 2026
Language
EN
Document type
Preprint