Developed Python Scripts for Propensity Score Matching

📅 2023-02-15 — Session: Developed Python Scripts for Propensity Score Matching

🕒 14:10–16:15
🏷️ Labels: Python, Propensity Score Matching, Infrastructure Investment, Data Analysis
📂 Project: Dev
⭐ Priority: MEDIUM

Session Goal:

The session aimed to develop and refine Python scripts for implementing Propensity Score Matching (PSM) in infrastructure investment studies.

Key Activities:

• Reviewed literature on infrastructure investment and economic growth, focusing on methodologies like propensity score matching (PSM).
• Expanded the methods section for evaluating infrastructure investment impacts, detailing sample selection, outcome variables, covariates, and regression analysis.
• Developed Python code snippets for estimating treatment effects using PSM, including logistic regression for propensity score estimation and nearest neighbor matching.
• Addressed an undefined variable error in PSM code and provided a corrected version.
• Discussed the importance of incorporating confounding variables in PSM models to enhance accuracy.

Achievements:

• Successfully created Python scripts for PSM, including data handling, model fitting, and treatment effect estimation.
• Clarified the role of confounders in PSM models and their impact on treatment effect accuracy.

Pending Tasks:

• Further validation and testing of the developed scripts using real-world datasets.
• Exploration of additional Python packages like causalml and sklearn for PSM implementation.