Wielding Occam's razor: Fast and frugal retail forecasting

TL;DR

This paper advocates for simpler, more efficient retail forecasting models by demonstrating that a reduced set of models can match complex methods in accuracy, while saving costs and reducing environmental impact.

Contribution

It introduces a framework for selecting parsimonious models that balance accuracy and computational cost without sacrificing forecast quality.

Findings

Reduced model sets perform comparably to complex methods.

Simpler models lead to significant cost savings.

Environmental benefits from decreased computational resources.

Abstract

The algorithms available for retail forecasting have increased in complexity. Newer methods, such as machine learning, are inherently complex. The more traditional families of forecasting models, such as exponential smoothing and autoregressive integrated moving averages, have expanded to contain multiple possible forms and forecasting profiles. We question complexity in forecasting and the need to consider such large families of models. Our argument is that parsimoniously identifying suitable subsets of models will not decrease forecasting accuracy nor will it reduce the ability to estimate forecast uncertainty. We propose a framework that balances forecasting performance versus computational cost, resulting in the consideration of only a reduced set of models. We empirically demonstrate that a reduced set performs well. Finally, we translate computational benefits to monetary cost savings and environmental impact and discuss the implications of our results in the context of large retailers.