Prediction of hierarchical time series using structured regularization and its application to artificial neural networks

07/30/2020
by   Tomokaze Shiratori, et al.
0

This paper discusses the prediction of hierarchical time series, where each upper-level time series is calculated by summing appropriate lower-level time series. Forecasts for such hierarchical time series should be coherent, meaning that the forecast for an upper-level time series equals the sum of forecasts for corresponding lower-level time series. Previous methods for making coherent forecasts consist of two phases: first computing base (incoherent) forecasts and then reconciling those forecasts based on their inherent hierarchical structure. With the aim of improving time series predictions, we propose a structured regularization method for completing both phases simultaneously. The proposed method is based on a prediction model for bottom-level time series and uses a structured regularization term to incorporate upper-level forecasts into the prediction model. We also develop a backpropagation algorithm specialized for application of our method to artificial neural networks for time series prediction. Experimental results using synthetic and real-world datasets demonstrate the superiority of our method in terms of prediction accuracy and computational efficiency.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
09/30/2022

Hierarchies Everywhere – Managing Measuring Uncertainty in Hierarchical Time Series

We examine the problem of making reconciled forecasts of large collectio...
research
04/08/2020

Fully reconciled GDP forecasts from Income and Expenditure sides

In this paper we re-consider the results of Athanasopoulos et al. (2019)...
research
04/21/2022

A Top-Down Approach to Hierarchically Coherent Probabilistic Forecasting

Hierarchical forecasting is a key problem in many practical multivariate...
research
06/07/2019

Reconciling Hierarchical Forecasts via Bayes' Rule

When time series are organized into hierarchies, the forecasts have to s...
research
12/01/2019

Machine learning applications in time series hierarchical forecasting

Hierarchical forecasting (HF) is needed in many situations in the supply...
research
01/05/2021

A Trainable Reconciliation Method for Hierarchical Time-Series

In numerous applications, it is required to produce forecasts for multip...
research
07/19/2022

Probabilistic Reconciliation of Count Time Series

We propose a principled method for the reconciliation of any probabilist...

Please sign up or login with your details

Forgot password? Click here to reset