Tipping-Point-Paper

This repository accompanies the research paper:

Sharp Hybrid Confidence Bands for Partially Identified Treatment Effects under Tail Uncertainty with an Application to Workforce Gender Diversity and Firm Performance Grace Lordan & Kaveh Salehzadeh-Nobari (2025)
https://arxiv.org/abs/2509.01622

It provides all datasets, code, and Monte-Carlo experiments required to replicate the empirical analysis and identification results in the paper.

---

What This Repository Provides

• Non-parametric identification of treatment effects at tipping points
• Support-based Manski bounds
• DKW-calibrated hybrid bounds for finite-sample validity
• Latent conditional expectation bounding
• Monte-Carlo validation across multiple data-generating processes
• End-to-end empirical replication of the paper’s figures & tables
• Publication-ready plots and LaTeX exports

---

Getting Started

1️⃣ Clone the repository

git clone https://github.com/kavehsn/Tipping-Points-Paper.git
cd Tipping-Points-Paper

2️⃣ Create & activate environment

conda env create -f tipping_environment.yml
conda activate tipping

3️⃣ Launch JupyterLab

jupyter lab