📅 2023-03-09 — Session: Conducted Regression and Causal Inference Analysis

🕒 21:00–22:50
🏷️ Labels: Regression Analysis, Causal Inference, Mock Data, Data Visualization, Violence Data
📂 Project: Dev
⭐ Priority: MEDIUM

Session Goal: The session aimed to conduct regression and causal inference analysis using mock datasets to explore biases, confounding variables, and the robustness of regression models.

Key Activities:

  • Developed a Python script for covariate analysis and visualization to compare focal and null areas.
  • Generated a mock dataset for regression analysis, including treatment and control groups.
  • Outlined a memo for documenting violence information outcomes, focusing on data quality and matching procedures.
  • Assessed the quality of a violence information dataset, detailing data cleaning and preprocessing steps.
  • Reviewed matching procedures and diagnosis notebooks to analyze the impact of infrastructure investment on violence outcomes.
  • Conducted numerical tests with mock data to evaluate matching and regression methods for causal inference.
  • Summarized findings from regression models for causal inference, highlighting biases and confounding variables.

Achievements:

  • Completed the development of scripts and notebooks for regression and causal inference analysis.
  • Documented the quality assessment of violence data and outlined further research steps.

Pending Tasks:

  • Further investigation into the implications of biases and confounding variables on regression models is needed.
  • Continue exploring the impact of infrastructure investments on violence outcomes.