A developed stock price forecasting model using support vector machine combined with metaheuristic algorithms

Armin Mahmoodi (), Leila Hashemi, Milad Jasemi, Soroush Mehraban, Jeremy Laliberté and Richard C. Millar
Additional contact information
Armin Mahmoodi: Islamic Azad University, Masjed Soleyman Branch
Leila Hashemi: Islamic Azad University, Masjed Soleyman Branch
Milad Jasemi: Islamic Azad University, Masjed Soleyman Branch
Soroush Mehraban: Islamic Azad University, Masjed Soleyman Branch
Jeremy Laliberté: Islamic Azad University, Masjed Soleyman Branch
Richard C. Millar: Islamic Azad University, Masjed Soleyman Branch

OPSEARCH, 2023, vol. 60, issue 1, No 3, 59-86

Abstract: Abstract Today the accurate prediction of stock price movement is one of the most effective tools for investors. This prediction becomes more challenging when the stock market is naturally chaotic and uncertain. These stock attributes prevent most forecasting models from valuable stock data. In this sense, In this research, our purpose is to use a suitable structure to predict the trading signals of the stock market with high accuracy. For this purpose, two models for the analysis of technical adaptation were used. It can be seen that suport vector machine (SVM) is used with particle swarm optimization (PSO) where PSO is used as a fast and accurate classification to search the problem-solving space and finally the results are compared with the performance of two other meta heuristic algorithm including the neural network and Cuckoo search algoritm (CS). Based on the result, we can say that all the new models are trustworthy in 6 days, however, SVM-PSO is better than basic research. The hit rate of SVM-PSO is 77.5%, but the hit rate of neural network (basic research) and SVM-CS are 71.2, 71.4 respectively. In this study, we examined the years 2013–2021, and if longer periods are used, it may now be possible to achieve more optimal results. The results show that the performance of SVM-PSO is superior to the performance of the SVM-CS and most importantly the feed-forward static neural network algorithm of the literature as the standard one. In this research, two approaches (raw-based and signal-based) have been developed to generate input data for the model while, we use historical data, unexpected events have not been determined. However, for comparison, the hit rate is considered the percentage of correct predictions for 16 days.

Keywords: Stock market predicting; Candlestick technical analysis; Neural network; Support vector machine; Particle swarm optimization (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s12597-022-00608-x Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:opsear:v:60:y:2023:i:1:d:10.1007_s12597-022-00608-x

Ordering information: This journal article can be ordered from
http://www.springer. ... search/journal/12597

DOI: 10.1007/s12597-022-00608-x

Access Statistics for this article

OPSEARCH is currently edited by Birendra Mandal

More articles in OPSEARCH from Springer, Operational Research Society of India
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().