Modeling Socioeconomic Ascent in Connecticut Census Tracts
by Jacob Dichter
March 13, 2025
Abstract
Predicting socioeconomic development of towns and urban regions is crucial for effective resource allocation and strategic decision-making by governments and businesses. Socioeconomic studies have traditionally relied on conventional statistical methods or qualitative analyses, which can be limited in their ability to capture the complex, multi-dimensional nature of development dynamics and adapt to evolving patterns in the data. Recent advancements in machine learning offer new possibilities for understanding and predicting socioeconomic changes. Ensemble learning methods and neural networks have shown promising results in various domains, sometimes combining multiple modal approaches to improve prediction accuracy and robustness. This study focuses on comparing the various statistical learning methods, from OLS to ensemble learning techniques like random forest and gradient boosting, in predicting socioeconomic ascent across Connecticut’s census tracts. The project involves collecting and preprocessing a comprehensive set of socioeconomic indicators, training and evaluating multiple statistical learning models, and conducting a comparative analysis to identify the most effective approach. By providing insights into the relative strengths of these techniques, this research aims to support data-driven decision-making in urban planning, private sector investment, resource allocation, and economic development initiatives in Connecticut.
Contents
A series of articles covers components of my M.S. project implementation incuding:
- Literature Review
- Data Ingestion / ETL Process
- Using PCA to Operationalize the Target Variable (SES)
- Modeling (Linear, traditional approaches)
- Ensemble Learning Implementation
- Model Validation & Discussion
Why does this matter? Who cares?
1. Real Estate Investors & Developers
They want to know which areas will rise in home prices and desirability soon.
Your hotspots signal good investment or development opportunities.
##2. Local Governments & Urban Planners
Hotspots indicate areas needing infrastructure, schools, or transportation upgrades to support growth.
Helps with efficient resource allocation and long-term planning.
##3. Businesses & Retailers
Hotspots mean growing consumer spending power and educated workforce—ideal for targeting expansions or new stores.
##4. Policy Makers & Community Organizations
Identifying upward-trending areas helps in designing targeted policies to sustain growth or address inequalities.
tags: