Multi Asset Price Time Series Prediction Based on LSTM, GRU, and MLP

DOI

10.2991/978-94-6239-648-7_59How to use a DOI?

Keywords

Time Series Prediction; Stock Prices; Cryptocurrency; Multi Asset Modeling

Abstract

The financial asset price has nonlinear, stochastic and multidimensional characteristics, making it difficult to use linear forecasting models. Therefore, in this paper, three neural networks are used—Multi Layered Perceptron (MLP), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU)—to achieve the joint prediction of the cryptocurrency and stock assets. Four representative asset data (Domino’s Pizza – DPZ; Bitcoin – BTC; Netflix – NFLX; Amazon – AMZN), each containing 1,520 daily price observations, are used. Here a sliding window of 60 consecutive closing prices are used to predict the next price. To remove forward looking bias, the data are partitioned time-wise, where the first 80 percent serves as a training dataset while the last 20 percent serves as a test. The outcome shows that the predictions of the LSTM model are the most precise (the result is equal to GRU model), which is followed by the MLP model (which predict significantly worst). Besides, the precision of predicting stock assets is higher than the precision of the predicting cryptocurrency assets (the consequence of different volatility), confirming the adequacy of applying the recurrent neural networks for modeling of nonlinear and long-term dependence of the multi-asset financial series.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yaqi Liu
PY  - 2026
DA  - 2026/04/24
TI  - Multi Asset Price Time Series Prediction Based on LSTM, GRU, and MLP
BT  - Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
PB  - Atlantis Press
SP  - 533
EP  - 543
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-648-7_59
DO  - 10.2991/978-94-6239-648-7_59
ID  - Liu2026
ER  -