Temporal Fusion Transformer for Multi-Horizon Probabilistic Forecasting of Weekly Retail Sales

TL;DR

This paper introduces a Temporal Fusion Transformer model for weekly retail sales forecasting that combines static and dynamic data, producing accurate probabilistic predictions with interpretability, outperforming traditional models.

Contribution

The study presents a novel TFT architecture tailored for multi-horizon retail sales forecasting, integrating static and exogenous signals, and demonstrating superior performance and interpretability.

Findings

TFT achieves RMSE of $57.9k and R^2 of 0.9875 on Walmart data.

TFT outperforms XGB, CNN, LSTM, and CNN-LSTM baselines.

Model provides calibrated probabilistic forecasts with interpretability features.

Abstract

Accurate multi-horizon retail forecasts are critical for inventory and promotions. We present a novel study of weekly Walmart sales (45 stores, 2010--2012) using a Temporal Fusion Transformer (TFT) that fuses static store identifiers with time-varying exogenous signals (holidays, CPI, fuel price, temperature). The pipeline produces 1--5-week-ahead probabilistic forecasts via Quantile Loss, yielding calibrated 90\% prediction intervals and interpretability through variable-selection networks, static enrichment, and temporal attention. On a fixed 2012 hold-out dataset, TFT achieves an RMSE of \$57.9k USD per store-week and an $R^2$ of 0.9875. Across a 5-fold chronological cross-validation, the averages are RMSE = \$64.6k USD and $R^2$ = 0.9844, outperforming the XGB, CNN, LSTM, and CNN-LSTM baseline models. These results demonstrate practical value for inventory planning and holiday-period optimization, while maintaining model transparency.