Deep State Space Recurrent Neural Networks for Time Series Forecasting

TL;DR

This paper introduces a novel neural network framework combining econometric state space models with advanced RNN architectures like LSTM, GRU, and TKANs to improve time series forecasting in the cryptocurrency market, showing promising results especially with TKANs.

Contribution

It presents a new hybrid neural network framework that integrates state space models with RNNs, specifically applying and comparing LSTM, GRU, and TKANs for cryptocurrency market prediction.

Findings

TKANs show promising forecasting performance.

Deep neural networks outperform linear models.

State space RNNs effectively model complex market dynamics.

Abstract

We explore various neural network architectures for modeling the dynamics of the cryptocurrency market. Traditional linear models often fall short in accurately capturing the unique and complex dynamics of this market. In contrast, Deep Neural Networks (DNNs) have demonstrated considerable proficiency in time series forecasting. This papers introduces novel neural network framework that blend the principles of econometric state space models with the dynamic capabilities of Recurrent Neural Networks (RNNs). We propose state space models using Long Short Term Memory (LSTM), Gated Residual Units (GRU) and Temporal Kolmogorov-Arnold Networks (TKANs). According to the results, TKANs, inspired by Kolmogorov-Arnold Networks (KANs) and LSTM, demonstrate promising outcomes.