Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Feb 26, 2026

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

2.0K

A deep learning framework for financial time series using stacked autoencoders and long-short term memory.

Wei Bao1, Jun Yue2, Yulei Rao1

  • 1Business School, Central South University, Changsha, China.

Plos One
|July 15, 2017
PubMed
Summary

This study introduces a new deep learning model combining wavelet transforms (WT), stacked autoencoders (SAEs), and long-short term memory (LSTM) for improved stock price forecasting. The novel framework enhances predictive accuracy and profitability in financial markets.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

History of gestational diabetes, modifiable lifestyle factors, and risk of cardiovascular disease and mortality: a prospective cohort study.

American journal of preventive cardiology·2026
Same author

LINC00324 suppresses abnormal proliferation and promotes apoptosis of leukemia cells through the miR-10b-5p/PTEN pathway.

Neoplasma·2026
Same author

Large extracellular vesicles and blebbisomes in cancer: emerging and translational opportunities highlights.

Cell communication and signaling : CCS·2026
Same author

Comparative Effectiveness of Different Immobilization Techniques in Managing Reduction of Acute Anterior Shoulder Dislocation: A Retrospective Cohort Study.

Indian journal of orthopaedics·2026
Same author

Energy-replenishing, mitochondria-targeted hydrogel microspheres mitigate sarcopenia via cellular senescence amelioration.

Journal of controlled release : official journal of the Controlled Release Society·2026
Same author

Dietary inflammatory potential and dietary quality in relation to advanced cardiovascular-kidney-metabolic syndrome and mortality risk: traditional and machine learning-based analysis.

Nutrition journal·2026

Area of Science:

  • Computational finance
  • Artificial intelligence in finance
  • Time series analysis

Background:

  • Deep learning is increasingly applied in finance, attracting investor and researcher interest.
  • Accurate stock price forecasting remains a significant challenge due to market noise and complexity.

Purpose of the Study:

  • To propose a novel deep learning framework for stock price forecasting.
  • To integrate wavelet transforms (WT), stacked autoencoders (SAEs), and long-short term memory (LSTM) for enhanced prediction.
  • To introduce SAEs for hierarchical feature extraction in stock price prediction for the first time.

Main Methods:

  • A three-stage deep learning framework was developed.
  • Wavelet transforms (WT) were used to denoise stock price time series data.

Related Experiment Videos

Last Updated: Feb 26, 2026

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

2.0K
  • Stacked autoencoders (SAEs) generated deep, high-level features from denoised data.
  • Long-short term memory (LSTM) networks forecasted the next day's closing price using these features.

    • Main Results:

    • The proposed WT-SAEs-LSTM model was evaluated on six market indices and their futures.
    • The model demonstrated superior predictive accuracy compared to existing methods.
    • The framework also exhibited enhanced profitability performance.

    Conclusions:

    • The combined WT-SAEs-LSTM deep learning framework offers a powerful approach for stock price forecasting.
    • The novel integration of SAEs for feature extraction significantly improves prediction.
    • The model shows practical viability for investors seeking improved market performance.