A beautiful shock? Exploring the impact of pandemic shocks on the accuracy of AI forecasting in the beauty care industry

Abstract

This research focuses on the profound impact of the shocks caused by the COVID-19 pandemic on the accuracy of AI-based demand forecasting in the beauty care industry. It aims to understand the key factors that led to decreased forecasting accuracy during the pandemic and employs causal mediation analysis to systematically investigate this complex issue. The empirical analysis is conducted using extensive order data from a major beauty care product manufacturer and distributor, covering the pre-pandemic, pandemic, and post-pandemic periods. The findings reveal that it is primarily the increase in demand volatility, and not the surge in sales volume, that has led to an increase in forecasting errors. This research provides crucial insights into the nuanced effects of macroeconomic shocks and consumer behavior changes on AI-based forecasting within the beauty care industry. Furthermore, it highlights the importance of understanding the underlying mechanisms that drive forecasting errors, paving the way for more resilient and robust demand forecasting systems in the future.