🛍️ Rossmann Store Sales Forecasting

📌 Overview

Forecasts daily sales for 1,115 Rossmann stores using XGBoost. Helps optimize:

• Inventory management
• Staff scheduling
• Promotional planning

---

🏆 Key Results

✅ 17% MAPE (42% better than naive baseline)
✅ Processes 1M+ records in <5 minutes
✅ Identified 3 underperforming promo strategies

---

📂 Dataset

• Source: Rossmann Sales Forecasting on Kaggle
• train.csv: Historical sales data for 1,115 Rossmann stores
• store.csv: Additional information about each store (type, competition, promos)

---

🧠 Features Engineered

Feature	Description
WeekOfYear	Seasonality indicator
MonthsSinceCompetition	Time since a competitor opened near the store
IsPromoMonth	Detects effective promo timing
PrevWeekSales	Lag feature to capture recent sales trend

---

⚙️ Tools & Libraries

• Python 🐍
• Pandas & NumPy
• XGBoost (regression)
• Matplotlib
• Scikit-learn

---

🚀 Model Summary

• Model: XGBoost Regressor
• Evaluation Metric: MAPE (Mean Absolute Percentage Error)