Product Lifecycle De-trending for Sales Forecasting

Albert Lechner, Steve Gunn

Abstract

This work introduces a new way to improve the sales forecasting accuracy of time series models using product’s life cycle information. Most time series forecasts utilize historic data for forecasting because there is no data available for the future. The proposed approach should change this process and utilize product life cycle specific data to obtain future information including product life cycle changes. Therefore a decision tree regression was used to predict the shape parameters of the bass curve, which reflects a product’s life cycle over time. This curve is used in a consecutive step to de-trend the time series to exclude the underlying trend created through the age of a product. The sales forecasts accuracy was increased for all 11 years of a luxury car manufacturer, comparing the newly developed product life cycle de-trending approach to a common de-trending by differencing approach in a seasonal autoregressive integrated moving average framework.

Download


Paper Citation


in Harvard Style

Lechner A. and Gunn S. (2020). Product Lifecycle De-trending for Sales Forecasting.In Proceedings of the 5th International Conference on Complexity, Future Information Systems and Risk - Volume 1: COMPLEXIS, ISBN 978-989-758-427-5, pages 25-33. DOI: 10.5220/0009324300250033


in Bibtex Style

@conference{complexis20,
author={Albert Lechner and Steve Gunn},
title={Product Lifecycle De-trending for Sales Forecasting},
booktitle={Proceedings of the 5th International Conference on Complexity, Future Information Systems and Risk - Volume 1: COMPLEXIS,},
year={2020},
pages={25-33},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0009324300250033},
isbn={978-989-758-427-5},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 5th International Conference on Complexity, Future Information Systems and Risk - Volume 1: COMPLEXIS,
TI - Product Lifecycle De-trending for Sales Forecasting
SN - 978-989-758-427-5
AU - Lechner A.
AU - Gunn S.
PY - 2020
SP - 25
EP - 33
DO - 10.5220/0009324300250033