Stock prediction based on Bayesian-LSTM

Biao Huang, Qiao Ding, Guozi Sun, Huakang Li*

*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

22 Citations (Scopus)

Abstract

Fluctuations in stock market represent the changes in national economic objectively. Some machine learning algorithms, such as linear fitting and sequence mining, are employed to predict the stock market. However, linear fitting faces issue of over-fitting and black relationships with historical data, while sequence mining is short in efficiency and lack dynamic adaptations. This paper builds a modified Bayesian-LSTM (B-LSTM) model for stock prediction. Six indicators of the Chinese stock market in every day are the basic input for LSTM. In order to represent the economic wave, we defined a data set unit by week which means the basic unit in LSTM is data in one week. Furthermore, a Bayesian optimization model is proposed to estimate the unit number dynamically in different economic cycle. The experimental results demonstrated that the B-LSTM increase over 25% prediction than the conventional LSTM.

Original languageEnglish
Title of host publicationProceedingsof 2018 10th International Conference on Machine Learning and Computing, ICMLC 2018
PublisherAssociation for Computing Machinery
Pages128-133
Number of pages6
ISBN (Electronic)9781450363532
DOIs
Publication statusPublished - 26 Feb 2018
Externally publishedYes
Event10th International Conference on Machine Learning and Computing, ICMLC 2018 - Macau, China
Duration: 26 Feb 201828 Feb 2018

Publication series

NameACM International Conference Proceeding Series

Conference

Conference10th International Conference on Machine Learning and Computing, ICMLC 2018
Country/TerritoryChina
CityMacau
Period26/02/1828/02/18

Keywords

  • B-LSTM
  • Bayesian optimization
  • Data set unit
  • Stock market
  • Stock prediction

Cite this