Mugi Ganaa

This project examines the relationship between U.S. trade policy changes, particularly tariffs, and consumer price movements using data driven and AI based analytical methods. Publicly available economic and trade datasets were collected and cleaned to analyze trends in consumer prices across selected product categories before and after policy changes. Exploratory data analysis and machine learning models were applied to identify patterns, measure price sensitivity, and assess the potential impact of trade policies on consumers. The findings provide insight into how trade policy decisions may influence inflation and consumer costs, demonstrating the value of AI driven analysis for economic and policy evaluation.

This project studies whether geopolitical events like wars and conflicts affect everyday consumer prices in the United States. Using 313 monthly observations from 2000 to 2026, I combined data from BLS CPI, FRED macroeconomic variables, and the Geopolitical Risk (GPR) index. I built and tested four machine learning models LagLinear, GPRLinear, GPRBoosting, and a Stacking Ensemble. I found that geopolitical risk does not raise inflation directly. Instead, shocks travel through oil prices first, then gas prices, and CPI adjusts 3 to 6 months later. The simplest model, LagLinear, achieved the best results with a CPI R² of 0.994 and average prediction error of only 0.18%. Transportation prices were the most sensitive category to geopolitical events. An interactive web dashboard was also built and deployed using Streamlit.