We are currently using Prophet by Facebook which is an open source forecasting model they released in 2017. After testing our own ARIMA and other statistical models we found Prophet to be the most accurate. 

Here's a quote about the characteristics which are very similar to eCommerce:

"Prophet is optimized for the business forecast tasks we have encountered at Facebook, which typically have any of the following characteristics:

  • hourly, daily, or weekly observations with at least a few months (preferably a year) of history
  • strong multiple “human-scale” seasonalities: day of week and time of year
  • important holidays that occur at irregular intervals that are known in advance (e.g. the Super Bowl)
  • a reasonable number of missing observations or large outliers
  • historical trend changes, for instance due to product launches or logging changes
  • trends that are non-linear growth curves, where a trend hits a natural limit or saturates"

The forecasting model is under constant iteration and improvement and we strongly encourage feedback from you on the accuracy and/or tweaking that you would like to see!

Did this answer your question?