A Worrying Analysis of Probabilistic Time-series Models for Sales Forecasting
This research is significant for practitioners in sales forecasting, as it challenges the assumption that complex probabilistic models are superior for point estimation, suggesting simpler baselines may be more effective.
This paper analyzes three prominent probabilistic time-series models for sales forecasting on a large-scale dataset. The study found that simpler models like Multi-layer Perceptron and Linear Regression consistently outperformed the probabilistic models in point estimation metrics such as RMSE and MAPE, without requiring feature engineering.
Probabilistic time-series models become popular in the forecasting field as they help to make optimal decisions under uncertainty. Despite the growing interest, a lack of thorough analysis hinders choosing what is worth applying for the desired task. In this paper, we analyze the performance of three prominent probabilistic time-series models for sales forecasting. To remove the role of random chance in architecture's performance, we make two experimental principles; 1) Large-scale dataset with various cross-validation sets. 2) A standardized training and hyperparameter selection. The experimental results show that a simple Multi-layer Perceptron and Linear Regression outperform the probabilistic models on RMSE without any feature engineering. Overall, the probabilistic models fail to achieve better performance on point estimation, such as RMSE and MAPE, than comparably simple baselines. We analyze and discuss the performances of probabilistic time-series models.