Forecasting Banknote Flows in BdI Branches: Speed-up with Machine Learning
Author(s)
Date Issued
2020
Type
conferenceObject
Start Page
945
End Page
950
Abstract
Abstract Every day banknotes are withdrawn from and deposited to Banca d’Italia
branches by banks. Up to 455 time series need to be predicted simultaneously. In
order to find a model with the best trade-off between forecasting performance and
time consumption, a comparison between different machine learning approaches is
presented.
