Randomized Neural Networks for Forecasting Time Series with Multiple Seasonality

TL;DR

This paper introduces a novel randomization-based neural forecasting method that effectively handles multiple seasonality and nonstationarity in time series, offering fast training and high accuracy.

Contribution

It presents a new randomization-based learning approach for neural networks tailored to complex, multi-seasonal time series forecasting.

Findings

Competitive forecasting accuracy with fully-trained networks

Fast and easy training process

Effective handling of nonstationarity and multiple seasonality

Abstract

This work contributes to the development of neural forecasting models with novel randomization-based learning methods. These methods improve the fitting abilities of the neural model, in comparison to the standard method, by generating network parameters in accordance with the data and target function features. A pattern-based representation of time series makes the proposed approach useful for forecasting time series with multiple seasonality. In the simulation study, we evaluate the performance of the proposed models and find that they can compete in terms of forecasting accuracy with fully-trained networks. Extremely fast and easy training, simple architecture, ease of implementation, high accuracy as well as dealing with nonstationarity and multiple seasonality in time series make the proposed model very attractive for a wide range of complex time series forecasting problems.