What Dhanush demonstrated

Built an end-to-end demand forecasting workflow from 9,648 historical transactions: explored sales trends by region and product, decomposed the monthly series, and compared a moving-average baseline, linear regression, and SARIMA using a proper chronological train/test split. Proved SARIMA was the strongest model with lower MAE, RMSE, and MAPE, and translated the results into a concise executive recommendation for planning.

• Leakage-Free Time-Series Splitting
• Comparative Forecast Evaluation
• Seasonality and Trend Diagnosis