43
ICI Bucharest © Copyright 2012-2021. All rights reserved
ISSN: 1220-1766 eISSN: 1841-429X Studies in Informatics and Control, 30(2) 43-54, June 2021
https://doi.org/10.24846/v30i2y202104
1. Introduction
A simplifed defnition of the bourse or stock
exchange is: the market where diferent shares of
stock, bonds and other fnancial instruments are
being sold and bought. The stock market facilitates
the money fow between investors and stock issuers.
The moment when an investor believes that a certain
company can potentially develop, she/he buys stock
shares in order (a) to be part of that business or (b) to
sell the stock shares at a higher price. An unwritten
capital market law states: “Buy low, sell high”.
An investor must have a strategy when deciding
what needs to be bought/sold and also tactics for
buying/selling. In what regards the strategy, an
investor would buy stock shares if the issuer has
good fnancial results; has clients/outlets for his
products/services; gives dividends to shareholders;
the executive management has achieved good
results in recent years; the sector in which he
operates is expanding. Obviously when the issuer
is no longer satisfying these requirements the
investor would consider selling.
The stock market produces massive amounts of data.
This data needs to be processed by machine learning
(ML) techniques. In recent years more and more
artifcial intelligence algorithms have been utilized
to predict the stock market. A hybrid stock selection
method that included stock prediction using extreme
learning machine and stock scoring was applied on
the A-share market of China (Yang et al., 2019).
Switching regime, ANFIS and GARCH techniques
have been employed in order to design a forecasting
model for the stock market risk (Kristjanpoller &
Mitchell, 2018). A combined fuzzy system and
GARCH model was utilized to forecast stock market
volatility (Hung, 2011). A hybrid model that uses
neural networks (NNs) and fuzzy interference was
used as a forecasting system (Badrul et al., 2015).
Neural networks were yet again used to predict the
stock market index in (Belciug & Săndiță, 2017),
and (Moghaddam, Moghaddam & Esfandyari,
2016). A genetic algorithm-based approach to
feature discretization in artifcial neural network has
been used for the forecast of the stock price index
(Kim & Han, 2000). In (Pehlivanli, Asikgil & Gulay,
2016) the support vector machines (SVMs) are
combined with four flter methods based on di ferent
metrics to obtain fltered features for forecasting
stock prices for Istanbul Stock Exchange market.
A hybrid model that consists of two linear models
(autoregressive moving average model, exponential
smoothing model) and a recurrent neural network
were used for the prediction of stock returns (Rather,
Agarwal & Sastry, 2015). An interesting approach
was the analysis of twitter feeds that were found
to be correlated with the Dow Jones Industrial
Average (Bollen, Mao & Zeng, 2011). A hybrid
time-series model was used for forecasting leading
Competitive / Collaborative Statistical Learning
Framework for Forecasting Intraday
Stock Market Prices: A Case Study
Smaranda BELCIUG
1
, Adrian SĂNDIȚĂ
1
, Hariton COSTIN
2
*,
Silviu-Ioan BEJINARIU
2
, Pericle Gabriel MATEI
3
1
University of Craiova, 13 Alexandru Ioan Cuza St., Craiova, 200585, Romania
sbelciug@inf.ucv.ro
2
Institute of Computer Science, Romanian Academy, No. 2 Codrescu Street, Iași, 700481, Romania
hcostin@gmail.com (*Corresponding author)
3
Ferdinand I Military Technical Academy Bucharest, 39-49 George Coşbuc Blvd., Bucharest 5, 050141, Romania
pericle.matei@mta.ro
Abstract: This paper presents an intelligent decision system based on statistical learning that regards the tactics of an
investor in predicting the next intraday stock price. Signifcant percentages can be won or lost depending on the tactics
applied for buying/selling shares. This paper includes a case study regarding the efciency of a group of machine learning
techniques that work together in a competitive/collaborative manner with a view to achieving an overall price forecast for the
next intraday transaction. In order to illustrate the advantages of this intelligent decision system this work provides a concrete
example concerning the price forecast for the next intraday transaction for Transilvania Bank (TLV), the stock market at the
Bucharest Stock Exchange (BVB), Romania. An important part of the decision system lies in the competitive stage, because
only the best competitors are chosen for the ultimate decision-making process. In the collaborative stage of the statistical
learning framework one uses a weighted voting system that outputs the fnal intraday stock price. The results obtained show
that this intelligent system outperforms each stand-alone method.
Keywords: Artifcial intelligence, Machine learning, Prediction methods, Statistical learning, Stock markets.