Back to news

Short-term retail forecasts under the presence of promotions

28 May 2013

The Lancaster Centre for Forecasting is continuously engaging in retail forecasting and in particular the effects of retail promotions.

In a recent research study, (visiting fellow with a Lancaster PhD), and have been studying how best to produce short-term SKU level forecasts across the retail chain using weekly data. These forecasts take into account price and marketing information both for the promoted SKUs and its competition within the category. But for many categories that’s an awful lot of variables, unmanageably large in fact. What Huang et al. show is that advanced techniques improve accuracy across the board and that where competitive information is available it is only of limited value.

The LCF is always keen to further its collaboration with industry. If you are interested for more information on this work please contact Professor Robert Fildes.


Huang, T., Fildes, R., & Soopramanien, D. (2013). The value of competitive information in forecasting FMCG retail product sales and the variable selection problem. Working paper 2013:1. Lancaster, UK: Department of Management Science. Download the working paper.