Comparison of hyperparameter tuning optimization methods for LSTM and stock price prediction

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F70883521%3A28140%2F25%3A63599758" target="_blank" >RIV/70883521:28140/25:63599758 - isvavai.cz</a>

Result on the web

<a href="http://dx.doi.org/10.1007/978-3-031-84353-2_17" target="_blank" >http://dx.doi.org/10.1007/978-3-031-84353-2_17</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-84353-2_17" target="_blank" >10.1007/978-3-031-84353-2_17</a>

Result language

angličtina

Original language name

Comparison of hyperparameter tuning optimization methods for LSTM and stock price prediction

Original language description

This paper investigates using Long Short-Term Memory (LSTM) networks for predicting stock prices, focusing on major stocks like AAPL, MSFT, TSLA, META, and GOOG from 2016 to 2021. We employ several technical analysis indicators, such as moving averages, the relative strength index, and others, as inputs to our LSTM model. The study involves data preprocessing, optimization, and tuning of ten different hyperparameters to enhance the performance of LSTM model. A comparative analysis of optimization techniques, including standard random search, Bayesian, Nevergrad optimization, covariance matrix adaptation optimization, and other bio-inspired algorithms, shows variations in performance across most datasets. Performance is measured using typical statistical metrics like mean squared error, R2 score, and others, with results showing varying prediction accuracies among different stocks. The study highlights the critical influence of data quality on LSTM performance and suggests further research into optimal hyperparameter tuning for enhancing AI-driven financial analytics.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

S - Specificky vyzkum na vysokych skolach

Publication year

2025

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Lecture Notes in Artificial Intelligence

ISBN

978-3-031-84352-5

ISSN

2945-9133

e-ISSN

1611-3349

Number of pages

13

Pages from-to

195-208

Publisher name

Springer

Place of publication

Cham

Event location

Zakopane

Event date

Jun 16, 2024

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

001535042200017