A comprehensive statistical analysis using Multiple Linear Regression to predict GDP growth rates using macroeconomic indicators from the Jordà-Schularick-Taylor Macrohistory Database.
This project develops a robust predictive model for GDP growth using historical macroeconomic data spanning multiple countries. Through rigorous statistical analysis including stepwise selection, LASSO regularization, best subset selection, and comprehensive diagnostics, we identify the key drivers of economic growth and achieve 88-89% explanatory power (Adjusted R²).
- High Predictive Accuracy: Explains 88-89% of GDP growth variation
- Advanced Variable Selection: LASSO, Best Subset, and Stepwise methods
- Diagnostics: VIF analysis, Cook's distance, residual testing
- Real-World Applications: Policy analysis, business planning, investment strategy
- Identify robust and parsimonious combination of economic factors influencing GDP growth
- Develop a statistically significant and reliable predictive model
- Provide actionable insights for policymakers, businesses, and investors
Clone the repository
git clone https://github.com/cool51/gdp-growth-macro-regression-analysis.git
cd gdp-growth-macro-regression-analysisJordà-Schularick-Taylor (JST) Macrohistory Database
- Comprehensive macroeconomic and financial data
- Source: https://www.macrohistory.net/database
Òscar Jordà, Moritz Schularick, and Alan M. Taylor. 2017. "Macrofinancial History and the New Business Cycle Facts." In NBER Macroeconomics Annual 2016, volume 31, edited by Martin Eichenbaum and Jonathan A. Parker. Chicago: University of Chicago Press.
Mukul Atreya
M.S. Statistics & Data Science
Sam Houston State University
This is an academic project, but suggestions and feedback are welcome!
This project is licensed under the MIT License.
⭐ If you found this project useful, please consider giving it a star! ⭐