KAN vs LSTM Performance in Time Series Forecasting

TL;DR

This study compares KAN and LSTM models for stock price forecasting, finding LSTM significantly more accurate but KAN offers interpretability and efficiency advantages, guiding model choice based on application needs.

Contribution

It provides a comparative analysis of KAN and LSTM for time series forecasting, highlighting LSTM's superior accuracy and KAN's interpretability benefits.

Findings

LSTM outperforms KAN in predictive accuracy across all horizons.

KAN offers better interpretability and computational efficiency.

LSTM is recommended for accuracy-critical financial forecasting.

Abstract

This paper compares Kolmogorov-Arnold Networks (KAN) and Long Short-Term Memory networks (LSTM) for forecasting non-deterministic stock price data, evaluating predictive accuracy versus interpretability trade-offs using Root Mean Square Error (RMSE).LSTM demonstrates substantial superiority across all tested prediction horizons, confirming their established effectiveness for sequential data modelling. Standard KAN, while offering theoretical interpretability through the Kolmogorov-Arnold representation theorem, exhibits significantly higher error rates and limited practical applicability for time series forecasting. The results confirm LSTM dominance in accuracy-critical time series applications while identifying computational efficiency as KANs' primary advantage in resource-constrained scenarios where accuracy requirements are less stringent. The findings support LSTM adoption for practical financial forecasting while suggesting that continued research into specialised KAN architectures may yield future improvements.