This project aimed to develop a comprehensive forecasting model for Walmart’s weekly sales, leveraging three major analytical approaches: time series models (SARIMA, Winter’s Method), panel data models (Fixed Effects, Random Effects, Mundlak Models), and machine learning (Random Forest, Boosted Trees).

Discover how advanced analytics transformed customer retention and revenue strategies for a winery. This project highlights actionable insights from segmentation, CLV analysis, and predictive modeling to optimize engagement and profitability.

Understanding traffic stops can provide crucial insights into patterns of law enforcement and potential biases in policing. The project dives into the traffic stop data from Rhode Island, spanning from January 2005 to December 2015, sourced from the Stanford Open Policing Project. This detailed analysis reveals intriguing trends and disparities that call for further exploration and action.