Notesale: Turn your study into money

Document Preview

Extracts from the notes are below, to see the PDF you'll receive please use the links above

---

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies
Geogrpahic Information System (Nepal Open University)

StuDocu is not sponsored or endorsed by any college or university
Downloaded by sagar dhital (sagardhital729@gmail
...
Strader
Drake University, Troy
...
edu

John J
...
rozycki@drake
...
ROOT
DRAKE UNIV, TOM
...
EDU

Yu-Hsiang (John) Huang
Drake University, yu-hsiang
...
edu

Follow this and additional works at: https://scholarworks
...
csusb
...
; Rozycki, John J
...
; and Huang, Yu-Hsiang (John) (2020) "Machine
Learning Stock Market Prediction Studies: Review and Research Directions," Journal of International
Technology and Information Management: Vol
...
4 , Article 3
...
lib
...
edu/jitim/vol28/iss4/3

This Article is brought to you for free and open access by CSUSB ScholarWorks
...
For more information, please contact scholarworks@csusb
...


Downloaded by sagar dhital (sagardhital729@gmail
...
Strader
John J
...
Root
Yu-Hsiang (John) Huang
(Drake University)

ABSTRACT
Stock market investment strategies are complex and rely on an evaluation of vast
amounts of data
...
The objective for this study is to identify
directions for future machine learning stock market prediction research based upon
a review of current literature
...
Four categories emerge:
artificial neural network studies, support vector machine studies, studies using
genetic algorithms combined with other techniques, and studies using hybrid or
other artificial intelligence approaches
...
The final section provides overall conclusions and directions for
future research
...
In 2019, the value of
global equites surpassed $85 trillion (Pound, 2019)
...
In the past,
investors relied upon their personal experience to identify market patterns, but this

©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies

Strader et al
...
Simple statistical analysis of financial data provides some insights but,
in recent years, investment companies have increasingly used various forms of
artificial intelligence (AI) systems to look for patterns in massive amounts of realtime equity and economic data
...

The objective for this study is to identify directions for future machine learning
(ML) stock market prediction research based upon a review of current literature
...
This will provide artificial intelligence and finance researchers with
directions for future research into the use of ML techniques to predict stock market
index values and trends
...
The following are four highly cited articles that used this methodology
...
(2012) employed a systematic literature review to evaluate empirical
studies of ML models for software development effort estimation (SDEE)
...

The selected articles were published between 1991 and 2010
...
This literature review
methodology provided insights into the current state of SDEE research that was the
basis for researcher recommendations and guidelines for practitioners
...
Software fault
prediction is a process used in the early phases of the software development
lifecycle for detecting faulty software modules or classes
...
Each article was evaluated based
on their predictive performance and a comparison with other statistical and machine
learning techniques
...
The study concluded with a set of
practitioner guidelines and researcher recommendations
...
One study evaluated machine learning flood
prediction systems (Mosavi, Ozturk and Chau, 2018)
...
2017

64

Downloaded by sagar dhital (sagardhital729@gmail
...
These models were intended to
aid hydrologists in predicting floods in the short-term and long-term, and
identifying cost-effective solutions that minimize risk, loss of human life and
property damage
...
Based on the findings from
the review, the authors were able to provide guidelines for hydrologists and climate
scientists when choosing the best machine learning technique for a prediction task
...
Orthopedics is the area of medicine
focused on prevention, diagnosis, and treatment of bone and muscle disorders
...
They appraised each article based upon its
ML technique, orthopedic application, model data, and predictive performance
...

This present study utilizes a methodology that is similar to the ones described above
because it is also evaluating articles describing ML use for predictions in a highly
complex problem domain
...
A research framework (taxonomy) is then provided to
encompass each of the study categories and provide a description for how the
categories differ
...
This provides the
basis for making researcher recommendations
...
First, the process
used to identify relevant studies is described
...
The studies in each category are then individually summarized and
discussed to identify common findings, unique findings, limitations, and areas
where more study is needed
...


METHOD FOR IDENTIFYING RELEVANT STUDIES
Each researcher involved in this study conducted an independent search for peerreviewed journal articles where some form of machine learning was used to predict

©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies

Strader et al
...
Articles were found using Google Scholar,
EBSCO, and EconLit
...
Each study used one, or more, machine learning techniques to predict
stock market index values or expectations for whether the future index value will
rise or fall
...
Several studies were eliminated from the list because they only focused
on predicting individual stock values
...
al
...
Twentysix studies were included in the final list to provide a representative sample of
research studies in this area
...
Each researcher then reviewed each paper to identify
groups of related studies that used a single machine learning technique, or those
that used a hybrid or multi-method approach
...
Each of the articles fits into one of the
following four categories: (a) artificial neural network studies, (2) support vector
machine studies, (3) studies using genetic algorithms with other techniques, and (4)
studies using hybrid or other artificial intelligence approaches
...
Machine Learning Stock Market Prediction Study Research Taxonomy

In the following section, the individual articles included in each research taxonomy
category are summarized focusing on their unique model, dataset and contribution
...
A brief description
of each machine learning approach is also provided prior to describing the related
studies
...
2017

66

Downloaded by sagar dhital (sagardhital729@gmail
...
ANNs are computational
models based on biological neural networks
...

Signals are transmitted (propagated) through the connected nodes as they learn
based on examples and attempt to reduce the level of prediction error
...
The following provides a brief description of each
ANN-related study’s unique research focus and findings
...
The focus is on
trying to support profitable trading
...
The study uses the daily closing values of the Standard
and Poor’s 500 Index (S&P 500), the German DAX Index, the Japanese TOPIX
index, and London’s Financial Times Stock Exchange Index (FTSE All Share)
...
The sample for TOPIX covers the period from January 1,
1969 to November 11, 1999 since data from earlier years was not available
...

Enke and Thawornwong (2005) use a machine learning information gain technique
to evaluate the predictive relationships for numerous financial and economic
variables
...
A threshold is determined to select only the strongest
relevant variables to be retained in the forecasting models
...
A cross-validation technique is also employed
to improve the generalizability of several models
...
The results
show that the trading strategies guided by the classification models generate higher
risk-adjusted profits than the buy-and-hold strategy, the other neural network
models, and the linear regression models
...
2017

67

Downloaded by sagar dhital (sagardhital729@gmail
...


The next study introduces a stochastic time effective neural network model to
uncover the predictive relationships of numerous financial and economic variables
(Liao and Wang, 2010)
...
The nearer the historical data
time is to the present, the stronger the impact the data have on the predictive model
...
The data is from several stock markets including the Shanghai and Shenzhen
Stock Exchange Stock A Index (SAI), Stock B Index (SBI), and the Hang Seng
(HIS), Dow Jones Industrial Average (DJIA), NASDAQ Composite (IXIC) and
S&P500
...

Chavan and Patil (2013) contribute to our understanding of ANN stock
market prediction by surveying different model input parameters found in nine
published articles
...
Based on their survey, they find that
most ML techniques make use of technical variables instead of fundamental
variables for a particular stock price prediction, while microeconomic variables are
mostly used to predict stock market index values
...

Chong, Han and Park (2017) analyze deep learning networks for stock
market analysis and prediction
...
They provide an objective assessment of both the advantages and
drawbacks of deep learning algorithms for stock market analysis and prediction
...
Testing is done using data from 38
companies listed in the Korean KOSPI stock market from the period January 4,
2010 through December 30, 2014
...

Studies Using Support Vector Machines to Analyze Stock Markets
The second group of articles includes studies primarily using support vector
machines (SVMs) to make stock market predictions
...
2017

68

Downloaded by sagar dhital (sagardhital729@gmail
...
The technique uses supervised learning
...
An SVM model represents the
examples as points in a space with the goal of creating a gap between the categories
that is as wide as possible
...
For example, in the context of stock market
prediction, according to Schumaker and Chen (2010), SVM is a machine learning
algorithm that can classify a future stock price direction (rise or drop)
...
This
proposed hybrid feature selection method, named F-score and Supported Sequential
Forward Search (F_SSFS), combines the advantages of filter methods and wrapper
methods to select the optimal feature subset from the original feature set
...

The study focuses on predicting the direction of the NASDAQ index using
commodity, currency, and other financial market index data from November 8,
2001 through November 8, 2007
...
In addition, experimental results show that the proposed
SVM-based model combined with F_SSFS has the highest level of predictive
accuracy and generalization in comparison with the other three feature selection
methods
...
They
developed a predictive machine learning approach for financial news article
analysis using several different textual representations: Bag of Words, Noun
Phrases, and Named Entities
...

They estimated a discrete stock price twenty minutes after a news article was
released
...

Yeh, Huang and Lee (2011) address problems that arise when using support
vector regression to forecast stock market values when dealing with kernel function
hyperparameters
...
In their system, advantages from different

©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies

Strader et al
...
They develop a two-stage multiple-kernel learning algorithm by
incorporating sequential minimal optimization and the gradient projection method
...

The daily stock closing price datasets used for training, validating, and testing the
model were from October 2002 through March 2005
...
Das and Padhy (2012) use two
machine learning techniques: backpropagation (BP) and SVM to predict future
prices in the Indian stock market
...
The implementation is carried out using
MATLAB and SVM Tools (LS-SVM Tool Box)
...
One alternative machine learning method that
has potential to do this is incorporating genetic algorithms (GAs) with either ANNs
or SVMs to reduce single technique limitations
...
The evolutionary process begins with a set
of randomly generated problem solutions
...
The solutions with higher
fitness are retained (survival of the fittest) and combined with other high fitness
solutions to create a new generation of solutions
...

This process continues until a certain number of generations has been created or the
population of solutions reaches a satisfactory fitness level
...

In the first study in this category by Kim and Han (2000), they propose a genetic
algorithm approach to feature discretization and the determination of connection
weights for artificial neural networks to predict the value of a stock price index
...
In most
of these studies, however, the GA is only used to improve the learning algorithm

©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Journal of International Technology and Information Management

Volume 28, Number 4

itself
...
The research data used in
this study is technical indicators and the direction of change in the daily Korea stock
price index (KOSPI) from January 1989 to December 1998
...

A study by Kim and Lee (2004) is also based on the same two machine learning
techniques used in the previous study
...
The GA is incorporated
to improve the learning and generalizability of ANNs for stock market prediction
...
Three ANN feature transformation methods are compared
...
The authors found that the experimental results
indicate that the proposed approach reduces the dimensionality of the feature space
and decreases irrelevant factors for stock market prediction
...
Their system is based on the idea of multiple
classifier combination where different classifiers attempt to solve the same problem
and then their decisions are combined to reduce estimation errors and improve
overall classification accuracy
...
This study proposes an approach that is capable of incorporating the
subjective problem-solving knowledge of humans into the results of quantitative
models
...
Genetic algorithms are used to combine classifiers stemming from
three sources – machine learning, experts, and users
...

Kim and Shin (2007) investigate the effectiveness of a hybrid artificial
neural network and genetic algorithm method for stock market prediction
...
To estimate the aspects of the ATNN
and TDNN design, a general method based on trial and error along with various
heuristics or statistical techniques is proposed
...
Research
data in this study come from the daily Korea Stock Price Index 200 (KOSPI 200)

©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies

Strader et al
...
The results show that the proposed
integrated approach produces better results than that of the standard ATNN, TDNN,
and the recurrent neural network (RNN)
...
In the proposed learning
paradigm, a genetic algorithm is first used to select input features for LSSVM
learning
...

Finally, the evolving LSSVM learning paradigm with the best feature subset,
optimal parameters, and a mixed kernel is used to predict stock market movement
direction in terms of historical data series
...
The entire data set of monthly values
covers the period from January 1926 to December 2005 with a total of 960
observations
...

A dynamic fuzzy model is proposed by Chiu and Chen (2009) in combination with
a SVM to explore stock market dynamics
...
A GA adjusts the
influential degree of each input variable dynamically
...
A multi-period experiment is designed to simulate
stock market volatility
...
To evaluate the performance of the new integrated model, they compare
it with traditional forecast methods
...

The experimental results show that the model is more accurate when compared with
alternative prediction methods
...
This final category
describes studies that have used other unique, or multi-method, artificial
intelligence techniques in this problem domain
...
Lee and Jo (1999) developed a candlestick
chart analysis expert system for predicting the best stock market timing
...
2017

72

Downloaded by sagar dhital (sagardhital729@gmail
...

Defined patterns are classified into five forms of price movements: falling, rising,
neutral, trend continuation, and trend-reversal patterns
...
Through experiments
using data from a sample of stocks listed in the Korean stock market from January
1992 to June 1997, it was shown that the developed knowledge base was time and
field-independent
...
O, Lee, Lee and Zhang
(2006) present a new stock trading method that incorporates dynamic asset
allocation in a reinforcement-learning framework
...

Formulating the MP in the reinforcement learning framework is achieved through
an environment and learning agent design
...
Experimental results using Korean stock
market (KOSPI) index data from 1998 to 2003 show that the proposed MP method
outperforms other fixed asset-allocation strategies and reduces the risks inherent for
local traders
...
According to the efficient market hypothesis, stock prices
should follow a random walk pattern meaning that the market should not be
predictable with more than about 50 percent accuracy
...
They used the Hurst exponent to select a
period with great predictability and found that the best period for analysis was from
June 4, 1969, to June 4, 1973 (1010 trading days)
...
Through appropriate model
collaboration, the resulting prediction accuracy was better than random at 65
percent
...

Individual data mining techniques have successfully generated accurate stock price
movement forecasts, but, over time, traders have realized that they need to use
multiple forecasting methods to gather better information about the future of the
stock market
...
The approaches include linear discriminant analysis
(LDA), quadratic discriminant analysis (QDA), K-nearest neighbor classification,
naïve Bayes based on kernel estimation, logit model, tree-based classification,

©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies

Strader et al
...
They examine the
daily change of closing prices in the Hang Seng index based on five predictors using
data from January 3, 2000 to December 29, 2006
...

Specifically, SVM is better than LS-SVM for in-sample prediction but LS-SVM is,
in turn, better than the SVM for the out-of-sample forecasts in terms of hit rate and
error rate criteria
...
Guresen, Kavakutlu and Daim (2011) evaluate the effectiveness of a
multi-layer perceptron (MLP), a dynamic artificial neural network (DAN2), and
hybrid neural networks which use generalized autoregressive conditional
heteroscedasticity (GARCH) to extract new input variables
...
One finding is that the simple MLP seems to be the best and most practical
ANN architecture
...
In a study by Dai, Wu and Lu (2012), a time series prediction
model that combines nonlinear independent component analysis (NLICA) and
neural networks is proposed for forecasting Asian stock markets
...
In the
proposed method, they first use NLICA to transform the input space composed of
original time series data into the feature space consisting of independent
components representing underlying information from the original data
...
To evaluate the performance of the proposed approach,
data from the Nikkei 225 closing index and Shanghai B-share closing index from
February 2, 2004 through March 3, 2009 are used as illustrative examples
...

The study by Patel, Shah, Thakkar and Kotecha (2015) compares four Indian stock
market prediction models: ANN, SVM, random forest, and naive-Bayes with two
approaches for model input
...
They assess the accuracy of each of the
prediction models for each of the two input approaches
...
2017

74

Downloaded by sagar dhital (sagardhital729@gmail
...
The
experimental results suggest that, for the first input data approach, random forest
outperforms the other three prediction models
...

Dash and Dash (2016) introduce a novel decision support system using a
computationally efficient functional link artificial neural network (CEFLANN) and
a rule set to more effectively generate trading decisions
...

The CEFLANN network used in the decision support system produces a set of
continuous trading signals by analyzing the nonlinear relationship that exists
between some popular technical indicators
...
This is a novel approach focused on profitable stock trading decisions
through integration of the learning ability of the CEFLANN neural network with
the technical analysis rules
...
Model training and testing are done using five years (January
4, 2010 to December 31, 2014) of historical stock index price values from the S&P
Bombay Stock Exchange Sensitive Index (BSE SENSEX) and S&P 500
...

Li et
...
(2016) present the design and architecture for a trading signal mining
platform that employs an extreme learning machine (ELM) to make stock price
predictions based on two data sources concurrently
...
The results show that (1) both
RBF ELM and RBF SVM achieve higher prediction accuracy and faster prediction
speed than BPNN, (2) the RBF ELM achieves similar accuracy with the RBF SVM,
and (3) the RBF ELM has faster prediction speed than the RBF SVM
...
Results show that the strategy with more accurate signals will be more
profitable with less risk
...
The BRT algorithm endogenously selects the predictor variables

©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies

Strader et al
...
The BRT algorithm also accounts for a potential nonlinear dependence of the forecast error on the predictor variables and for
interdependencies between the predictor variables
...
Their main
finding is that, given the set of predictor variables used in this study, the rational
expectations hypothesis (REH) cannot be rejected for short-term forecasts and that
there is evidence against the REH for longer term forecasts
...

Zhong and Enke (2019) present a process to predict the daily return direction of a
set of stocks
...
While controlling for overfitting, a pattern for
the classification accuracy of the DNNs is detected and demonstrated as the number
of the hidden layers increases gradually from 12 to 1000
...
The trading strategies guided by the DNN
classification process based on PCA-represented data perform slightly better than
the others tested, including a comparison against two standard benchmarks
...
The daily data is from 2518 trading days
between June 1, 2003 and May 31, 2013
...
Given
the ML-related systems, problem contexts, and findings described in each selected
article, and the taxonomy categories presented earlier, several conclusions can be
made about our current knowledge in this research area
...
This
is analogous to task-technology fit (Goodhue and Thompson, 1995) where system
performance is determined by the appropriate match between tasks and
technologies
...
Support vector machines best fit classification problems
such as determining whether the overall stock market index is forecast to rise or
fall
...
2017

76

Downloaded by sagar dhital (sagardhital729@gmail
...
While each study did illustrate that the methods can be
effectively applied, the single method applications do have limitations
...
The problem is that, at some point, the systems become so complex
that they are not useful in practice
...

The second conclusion from this review of past studies is that generalizability of
findings needs to be improved
...
Three enhancements can be made for the experimental
system assessment
...
These systems could also be tested in the same time period for US or
European markets
...
For example, would an approach accurately
predict market values in the US during the financial crisis of 2008-2009 and also
during the recent market growth period from 2018-2019? If systems are able to
predict market growth, are they also able to predict market contraction? Finally,
proposed methods could be used to evaluate predictive performance for stock
market indices that include only small firms vs
...
Are systems
effective under different risk and volatility environments? Any of these
experimental method enhancements will provide a stronger research and practice
contribution
...
Financial investment
theory needs to be a stronger driver underlying the ML systems’ inputs, algorithms,
and performance measures
...
Too many studies use techniques without
consideration of the vast amount of financial theory that has been developed over
the past centuries
...
At this point this rarely occurs so it is
impossible to find patterns where there is a mismatch between a particular stock
market prediction problem and a machine learning technique
...
If the best machine
learning stock market prediction technique is found, and all investors adopt this
system, the result is that no one is better off
...


©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies

Strader et al
...
, Locoro, A
...
(2018)
...
Frontiers in Bioengineering and Biotechnology, 6, 75
...
S
...
T
...
Parameters for stock market prediction
...

Chiu, D
...
, & Chen, P
...
(2009)
...
Expert Systems with Applications, 36(2), 1240-1248
...
, Han, C
...
C
...
Deep learning networks for stock market
analysis and prediction: Methodology, data representations, and case studies
...

Dai, W
...
Y
...
J
...
Combining nonlinear independent
component analysis and neural network for the prediction of Asian stock market
indexes
...

Das, S
...
, & Padhy, S
...
Support vector machines for prediction of futures
prices in Indian stock market
...

Dash, R
...
K
...
A hybrid stock trading framework integrating
technical analysis with machine learning techniques
...

Enke, D
...
(2005)
...
Expert Systems with applications, 29(4), 927940
...
L
...
L
...
Task-technology fit and individual
performance
...

Guresen, E
...
, & Daim, T
...
(2011)
...
Expert Systems with Applications, 38(8),
10389-10397
...
2017

78

Downloaded by sagar dhital (sagardhital729@gmail
...
H
...
Adaptation in natural and artificial systems: an introductory
analysis with applications to biology, control, and artificial intelligence
...

Jasic, T
...
(2004)
...
Applied Financial Economics,
14(4), 285-297
...
J
...
S
...
A hybrid approach based on neural networks and
genetic algorithms for detecting temporal patterns in stock markets
...

Kim, K
...
, & Han, I
...
Genetic algorithms approach to feature discretization
in artificial neural networks for the prediction of stock price index
...

Kim, K
...
, & Lee, W
...
(2004)
...
Neural computing & applications,
13(3), 255-260
...
J
...
H
...
(2006)
...
Expert Systems
with Applications, 31(2), 241-247
...
, Pandey, A
...
, & Darbari, M
...
A hybrid machine
learning system for stock market forecasting
...

Lee, K
...
, & Jo, G
...
(1999)
...
Expert systems with applications, 16(4), 357-364
...
C
...
Using support vector machine with a hybrid feature selection
method to the stock trend prediction
...

Li, X
...
, Wang, R
...
, Cao, J
...
& Deng, X
...

Empirical analysis: stock market prediction via extreme learning machine
...

Liao, Z
...
(2010)
...
Expert Systems with Applications, 37(1), 834-841
...
2017

79

Downloaded by sagar dhital (sagardhital729@gmail
...


Malhotra, R
...
A systematic review of machine learning techniques for
software fault prediction
...

Mosavi, A
...
, & Chau, K
...
(2018)
...
Water, 10(11), 1536
...
, Lee, J
...
W
...
(2006)
...
Information Sciences,
176(15), 2121-2147
...
, & Wang, H
...
Prediction of stock market index movement by ten data
mining techniques
...

Patel, J
...
, Thakkar, P
...
(2015)
...
Expert Systems with Applications, 42(1), 259-268
...
, & Risse, M
...
A machine‐learning analysis of the rationality
of aggregate stock market forecasts
...

Pound, J
...
Global stock markets gained $17 trillion in value
in 2019
...
cnbc
...
html
...
, & Rasheed, K
...
Stock market prediction with multiple classifiers
...

Schumaker, R
...
, & Chen, H
...
Textual analysis of stock market prediction
using breaking financial news: The AZFin text system
...

Schumaker, R
...
, & Chen, H
...
A discrete stock price prediction engine
based on financial news
...

Wen, J
...
, Lin, Z
...
, & Huang, C
...
Systematic literature review
of machine learning based software development effort estimation models
...


©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Journal of International Technology and Information Management

Volume 28, Number 4

Yeh, C
...
, Huang, C
...
, & Lee, S
...
(2011)
...
Expert Systems with
Applications, 38(3), 2177-2186
...
, Chen, H
...
, & Lai, K
...
(2008)
...
IEEE Transactions on evolutionary
computation, 13(1), 87-102
...
, & Enke, D
...
Predicting the daily return direction of the stock
market using hybrid machine learning algorithms
...


APPENDIX
List of Reviewed Machine Learning Stock Market Prediction
Articles
Author(s)
[publication year]
Jasic and Wood [2004]

Machine Learning Method(s)

Primary Market(s) Studied and
Data Time Period (Years)
S&P 500, DAX, TOPIX and
FTSE
1965-1999
S&P 500 1976-1999

Artificial neural network

Enke and Thawornwong Artificial neural network
[2005]
Liao and Wang [2010]
Artificial neural network

Shanghai and Shenzhen Stock
Exchange 1990-2008
Study reviews nine ANN studies
Korean KOSPI stock market
2010-2014

Chavan and Patil [2013]
Artificial neural network
Chong, Han and Park Artificial neural network
[2017]
Lee (2009)
Schumaker and Chen
[2009]
Yeh, Huang and Lee
[2011]
Das and Padhy [2012]

Support vector machine
Support vector machine

Kim and Han [2000]

Genetic algorithm with artificial neural Korea stock price index (KOSPI)
network
1989-1998
Genetic algorithm with artificial neural Korea composite stock price
network
index (KOSPI) 1991-1998

Kim and Lee (2004)

NASDAQ 2001-2007
Companies listed in the S&P500
in 2005
Taiwan Capitalization Weighted
Stock Index 2002-2005
National Stock Exchange (NSE)
of India Limited 2007-2010

Support vector machine
Support vector machine

©International Information Management Association, Inc
...
com)

ISSN: 1941-6679-On-line Copy

lOMoARcPSD|12620520

Machine Learning Stock Market Prediction Studies

Strader et al
...
al
...
2017

82

Downloaded by sagar dhital (sagardhital729@gmail
...
2017

83

Downloaded by sagar dhital (sagardhital729@gmail

---