📅 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.