STOCK PRICE PREDICTION USING FUNDAMENTAL

ANALYSIS WITH LSTM

A MAJOR PROJECT REPORT

Submitted by B449
AKSHAT UNIYAL [RegNo: RA1911028010061] &
RAHUL GOEL [RegNo: RA1911030010094]
Under the Guidance of

Dr. V. JOSEPH RAYMOND

(Assistant Professor , Department of Networking and Communication )

in partial fulfilment of the requirements for the degree

of

BACHELOR OF TECHNOLOGY
in
COMPUTER SCIENCE ENGINEERING
of
FACULTY OF ENGINEERING & TECHNOLOGY

SRM Nagar, Kattankulathur, Chengalpettu District

February 2023

1
 SRM INSTITUTE OF SCIENCE AND TECHNOLOGY
(Under Section 3 of UGC Act, 1956)

BONAFIDE CERTIFICATE

Certified that this Major project report titled “STOCK PRICE

PREDICTION USING FUNDAMENTAL ANALYSIS WITH
LSTM” is the bonafide work of Mr. Akshat Uniyal [Reg No:
RA1911028010061] & Mr. Rahul Goel [Reg No:
RA1911030010094], who carried out the project work under my
supervision. Certified further, that to the best of my knowledge the
work reported herein does not form part of any other thesis or
dissertation on the basis of which a degree or award was conferred on
an earlier occasion for this or any other candidate.

SIGNATURE SIGNATURE

Dr. V. Joseph Raymond Dr. Annapurani Panaiyappan

GUIDE HEAD OF THE DEPT.
Assistant Professor Professor
Dept. Of Computing Technologies Department Of Computing Technologies

2
 TABLE OF CONTENTS

Acknowledgement……………………………………………. 5

Abstract ……………………………………………………… 7

List Of Symbols and Abbreviations …………………………. 8

Introduction ………………………………………………….. 9

General ………………………………………………... 10

Problem Statement ……………………………………. 11

Objective ……………………………………………… 12

Data Analysis and Stock Prediction ………………….. 12

Introduction to The Stock Market ……………………. 13

Literature Survey …………………………………………… 15

Algorithms Description …………………………………….. 19

Long Short Term Memory …………………………… 22

Min Max Scaling ……………………………………... 22

System Analysis…………………………………………….. 23

Software Requirements ………………………………. 23

Hardware Requirements ……………………………… 23

System Implementation …………………………………….. 24

3
Methodology ………………………………………………. 28

Conclusion ………………………………………………… 29

References .………………………………………………… 29

Implementation ..…………………………………………... 31

4
 ACKNOWLEDGEMENT

I would like to express my sincere thanks to Dr. C.

Muthamizhchelvan, Vice Chancellor of Kattankulathur SRM Institute
of Science and Technology for extensive facilities for the work of the
project and his continued support.

Sincere thanks Dr T. V. Gopal, Dean-CET, SRM Institute of

Technology, Kattankulathur and Dr. Revathi Venkataraman,
Professor, School of Computing, SRM Institute of Technology, for
their support throughout the work on the project.

I would like to thank our Head of Department, Dr. K. Annapurani

Panaiyappan, Professor, Department of Networking and
Communications, SRM Faculty of Technology, for his guidance and
encouragement at various stages of the project work. Also, I would
like to thank our instructor Dr. Mr. Suchitra, Associate Professor,
Department of Computing Technology, SRM Institute of Technology,
supervised and assisted me throughout the course.

I would like to thank my guide, Dr. V. Joseph Raymond, Assistant

Professor, Department of Networking and Communications, SRM
College of Technology, who gave me the opportunity to pursue my
project under his guidance. It gave me the freedom and support to
explore the research topics that interested me. Her passion for solving
problems and making a difference in the world has been inspirational.

Finally, I would like to express my deepest regards to the teachers and

students of the Networks and Communication(NwC) Department of

5
SRM Institute of Technology for their help throughout my project.
Finally, I would like to thank my parents, family and friends for their
unconditional love, constant support and encouragement.

6
 ABSTRACT

Stock market forecasts are always attracting the attention of multiple

analysts and researchers. Many Popular theories observe that the
stock market is in it's essence a random game of chance and that it's a
mindless game to try to predict a thin item. Predicting the price of a
stock is in itself a challenging problem due to the number of external
variables present .

In the short term, the market acts like an election or a voting machine,
but in the long term it acts like a weighing machine and thus can
make predictions on market movements over a longer time frame.
Stock Price prediction’s integration with modern technology –
especially Machine Learning Algorithm (Quant models as they are
referred to in the financial sector) is recently becoming a growing
idea for research.

Research has shown that Machine Learning Models particularly with

the use of Recursive Neural Networks (RNN) and Long-Term Short
Memory (LSTM) when applied to historical data of stocks can be
applied to predict the short-term price of the share. Our model applies
companies’ projections as well as sector performances and how the
given company fares accordingly to correctly predict equity share
prices for both short- and long-term goals.

7
 LIST OF SYMBOLS AND ABREVIATIONS

S&P - Standard & Poors 500 (USA)

DJI - Dow Jones (USA)
NAS - National Association of Securities Dealers Automated
Quotations (USA)
Nikkei225 - Nikkei 225 Stock Average(Japan),
SSE - Shanghai Stock Exchange (Shanghai/China),
HSI - Hang Seng Index (Hong Kong),
BSE - Bombay Stock Exchange (India),
DAX - Data Analysis Expressions (Europe)
SMI - Swiss Market Index (Switzerland)
MXX - Mexico Stock Exchange (Mexico)
BVSP - Brazil's benchmark Bovespa stock (Brazil)

8
 CHAPTER 1

INTRODUCTION

1.1 GENERAL

Stock market forecasting and analysis involves determining the future

value of a company's stock or other exchange-traded financial
instrument. The stock market is an important part of the country's
economy, plays an important role in the growth of the country's
industry and trade, and ultimately affects the country's economy.
Investors and industries participate in the stock market and both want
to know if certain stocks will rise or fall over a certain period of time.
The stock market is the main source for any company to raise funds
for the expansion of its business. It is based on the concept of supply
and demand. If demand for a company's stock is high, the company's
stock price increases; if the demand for the company's stock is low,
the company's stock price goes down.

The National Stock Exchange of India Limited (NSE) is the main

Indian stock exchange, based in Mumbai. Established in 1992, NSE
was the country's first equity-based electronic stock exchange. NSE is
the first exchange in the country to offer a modern, fully automated

9
electronic screen trading system, providing convenient trading
facilities to investors nationwide.

The main motive to correctly predict stock values and prices in the
short and long term is to maximise your potential earnings rather than
relying on tips. A huge amount of research has gone into developing
Machine Learning models that are capable of correctly predicting
stock prices and have been used by hedge funds and investment banks
for quite some time now. However these models are mainly used to
predict short term prices so that they can be utilised in intraday
trading and most long term models generally focus on indices and
option chains.

The efficiency of various various prediction models can be debated as

many can’t predict long term fluctuations and compare the current
stock value as compared to its current trading price which takes into
account the sector performance(For E.g.: The existing share of a stock
let suppose Reliance declines in the Oil Sector as compared to its
projections signalling a down quarter but it’s price value doesn’t
fluctuate much leading to the assumption that a correction is in order
which will result in large volumes of shares being sold and the stock
price taking a hit).

Fundamental Analysis refers to concept of using underlying financial

records published by the company and taking other competitors’ data
and comparing them to correctly predict the short- and long-term
prices.

10
Due to the large number of industries and companies involved, it
contains very large datasets from which it is difficult to extract
information and manually analyse their work trends. The application
developed in this project not only helps in predicting the future
movement of stocks in the market, but also automates data retrieval,
trend analysis, predictive analysis and information generation for
stocks in one click.

Stock market analysis and prediction, revealing market shape,

predicting buying time. Successfully predicting the future price of a
stock can generate substantial profits. This was done using historical
market data spanning 12 months in this project to represent different
conditions and to confirm that time series models have statistically
significant predictive power in predicting high levels of competitive
trading investment Probability of trades profitable and high profitable
returns.

1.2 PROBLEM STATEMENT

The Stock Market is a complex and dynamical system, & is

influenced by many factors that are subject to uncertainty. So, it is a
difficult task to forecast any price movements. Due to technology and
globalization of business & financial markets it is important to
analyse the movement of share prices over short and long terms
respectively.

ML algorithms can be can be developed and trained based on

financial predictive indicators (like RSI, VIX, GOLD, etc.) and
algorithms for the analysis of stock prices in Indexes like in NSE’s
Nifty 50 Index, BANK NIFTY, S&P500 (USA), Dow Jones (USA),

11
Nasdaq (USA), Nikkei225 (Japan), SSE (Shanghai/China), HSI
(Hong Kong), BSESN (India), DAX (Europe), SMI (Switzerland),
MXX (Mexico), BVSP (Brazil).

1.3 OBJECTIVE

The main objective of the project is to add to the academic

understanding of the stock market. With a greater understanding of
how the market works and moves, investors will be better equipped to
prevent another financial crisis. They would be able to make Buy/Sell
decisions on their own with the help of their own analysis, which will
benefit them in the long run. It will help traders to manage their risk
better and not put their entire capital in jeopardy. It will lead to better
understanding of correlation between major Indexes of the world
along with the knowledge of advanced indicators like volume,
drawdowns, VIX, etc. and how to use them for the analysis of the
market.

1.4 DATA ANALYSIS (DA) AND STOCK

PREDICTION (SP)

Data analysis (DA) in machine learning (ML) is a process of applying

technical skills (ML Algorithms) on historical data to obtain statistical
as well as tabular results about predictions. It is also considered as a
technical process of data illustration and evaluation. Two authors
(Shamoo et al,2012; Resnik et al ,2013) explained about DA,
12
according to their theory DA is the process of distinguishing signals
for decision making with statistical fluctuation of results. DA also
included collection as well as analyzing process, it can be iterative
according to problem statement. Several statistical techniques
implemented in DA.

Data scientists find patterns of entire data with special observations.

Several types of quantitative as well as qualitative approaches such as
content analysis, history analysis, sentimental analysis and
bibliographic analysis are involved in DA. DA study formulate
predictions on the basis of historical data that can be present in form
of notes, files, documents, tables, audio or video tapes. Accurate
analysis of different research findings can lead to valid knowledge
discovery. Inaccurate statistical presentation of data destroys the
research findings of any scientist and guides wrong destinations to
readers and public perceptions about research also influenced
negatively. Integrity and accuracy are two main parts of statistical
data analysis. In this research we used several data resources in the
form of datasets, research theories and financial resources of data
presentation. Yahoo Finance, Quandle, Kaggle and several other
similar platforms provided data that is used in stock market
predictions. We obtained data from these platforms for different stock
exchange companies and after applying ML algorithms we presented
stocks predictions results statistically.

1.5 INTRODUCTION TO STOCK MARKET

Stock market can be defined as a combined platform of several

markets and exchangers with a regular process of buying and selling
13
goods that shares issued publically (Comparison analysis performed
at public platform). At this platform several situational financials
performed for the formal exchange process under defined rules and
regulations. Multiple stocks trading places can be available at
different places in a country where transactions on stocks can be
performed. There are two different terms used in stock market
definition as stock exchange and stock market with consideration of
formal trade assets. We can define the stock market as, if someone is
part of the stock market will be considered as a stock market trader, as
he buys or sells his shares on one or different stocks. Several stock
exchange markets available at national as well as international level
including Pakistan and world level biggest stock market i.e. New
York Stock Exchange (NYSE), Chicago (CBOE) both of these are
considered as national stock markets of the USA.

14
 CHAPTER 2

LITERATURE SURVEY

[1] Research on Legit Neural Network Based Stock Price Prediction

Method, IEEE 2019 – Sayavong Lounnapha et al. This paper aims to
build a stock price prediction model centered on a complex neural
network with excellent self-learning ability. Behavioral training and
testing datasets based on cumulative neural networks and Thai stock
price prediction. The prediction accuracy is high, which can also be
promoted in the financial field.

[2] Enhancing returns by predicting stock prices using Deep Neural

Networks, IEEE 2019-Soheila Abrishami, et al., Economic Time
Series Forecasting is an important company that has been collected,
has attracted the attention of many researchers, and is extremely
important for investors. This article focuses on a deep learning system
that predicts the value of a stock using a series of data from a subset
of stocks on the Nasdaq stock exchange. The model is trained on
minimal data for a specific stock and accurately estimates the final
value of that stock in just a few steps. It includes an auto-encoder to

15
remove noise and uses time series data engineering to provide
enhancements with original features. These new features are fed to a
stacked LSTM auto-encoder for a multi-step estimation of the final
stock value. Also, this estimate is used as part of the profit
maximization method to provide support at the right time to buy or
sell a particular stock. The results show that the proposed framework
outperforms state-of-the-art time series forecasting methods in terms
of analytical accuracy and efficiency.

[3] LSTM Method for Bitcoin Price Prediction: A Case Study Stock
Market Yahoo Finance, IEEE 2019- Ferdiansyah et al., The stock
market is very risky. Bitcoin is a cryptocurrency. In recent years, the
price has gradually increased, and sometimes it suddenly dropped. I
don't know if it will affect the scholarship. Due to its volatility,
automated tools are needed to predict Bitcoin in the stock market.
This study examines how LSTMs can be used to generate forecasts of
the Bitcoin stock market prediction model. Before confirming the
results, the article attempts to measure the results using RMSE (root
mean square error). RMSE will always be greater than or equal to
MAE. The RMSE metric evaluates the model's ability to compute
continuous values. The method applied in this study to predict Bitcoin
on the Yahoo Finance stock market can predict results above $12,600
within two days of the prediction.

[4] Stock Price Prediction Using Machine Learning Techniques, IEEE

2019-Jeevan B et al., Recently, the stock market has been in the
headlines and more and more scholars and companies are interested in
it. This article mainly discusses the use of RNN (Regenerative Neural
Network) and LSTM (Long Short-Term Memory) to predict stock
prices using various factors on national stock exchanges. Factors such
as current market prices as well as anonymous events. The article also

16
mentions recommender systems for company selection and models
built on the RNN and LSTM methods.

[5] Stock Market Prediction Using Machine Learning Techniques,

IEEE 2020- Naadun Sirimevan et al., Stock prices play an important
role in today's economy. Researchers found that social media
platforms such as Twitter and online news tend to influence any
individual's decision-making process. In this study, behavioral
responses to online news were found to close the gap and make
predictions more accurate. Accurate predictions were made a day, a
week, two weeks later.

[6] Stock Market Forecasting by Machine Learning, IEEE 2018 -

Ishita Parmar, Ridam Arora, Lokesh Chouhan, Navanshu Agarwal,
Shikhin Gupta, Sheirsh Saxena, Himanshu Dhiman. In this article, we
investigate the application of regression and LSTM-based machine
learning to stock price prediction. The items measured are open price,
close price, low price, high price and volume. This article attempts to
use machine learning techniques to determine a company's future
stock price with increased accuracy and reliability. The LSTM
algorithm gave more accurate positive results in predicting stock
prices.

[7] Stock Price Prediction Using Machine Learning, IEEE 2018 -

Jeevan B, Naresh E, Vijaya Kumar BP, Prashanth Kambli. This paper
is mainly based on the action process prediction method, using long-
short-term memory (LSTM) and recurrent neural network (RNN) to
predict action value. ESN data is analyzed using various factors such
as market price streams, price-earnings ratios, fundamental values and

17
other anonymous facts. Analyze model performance by comparing
actual and predicted data using an RNN plot. Machine learning
predicts stock prices because the model can predict stock prices very
close to the actual price, where the model captures detailed
characteristics and uses different prediction strategies. The template
starts all NSE internet data and identifies inputs and groups them and
provides inputs based on user configuration. This RNN-based
architecture has proven to be very good at predicting performance by
modifying configurations accordingly, which also uses
backpropagation during acquisition. and group the data to avoid data
mixing.

[8] Stock Market Forecasting Using Machine Learning, IEEE 2016 -

Mehak Usmani, Syed Hasan Adil, Kamran Raza, Syed Saad Azhar
Ali. The main objective of this study is to predict the performance of
the Karachi Stock Exchange (KSE) closing market using machine
learning algorithms. Various characteristics such as market entry and
positive and negative forecasts can be predicted using predictive
models. Features used in the model include oil prices, gold and silver
rates, interest rates, foreign exchange rates (FEX), news, and social
media feeds. Machine learning algorithms such as single-layer
perceptron (SLP), multi-layer perceptron (MLP), radial basis function
(RBF), and support vector machine (SVM) are compared. The MLP
algorithm is a multi-layer perceptron that achieves the best results
compared to the different methods. The most useful feature for
predicting the market is the oil price attribute. The final results of the
study confirmed that machine learning techniques can predict stock
market performance. The Perceptron multi-layered machine learning
algorithm correctly predicted 70% of market performance.

18
[9] Predictive Model Development for Stock Analysis, IEEE 2017 -
R. Yamini Nivetha, Dr. C. Dhaya. Related study of three algorithms
namely – Multiple Linear Regression (MLR), Support Vector
Machine (SVM) and Artificial Neural Network (ANN) is the main
focus of this study. Forecast the next day's market price, the forecast
is divided into monthly forecast and daily forecast. With monthly and
daily forecasts. Sentiment analysis using the best forecasting
algorithms can predict stock prices. The least developed algorithm is
the multiple linear regression algorithm which calculates the
correlation between volume and stock price. The research results
show that the deep learning algorithm is more advanced than the
MLR algorithm and the SVM algorithm.

19
 CHAPTER 3

ALGORITHMS DESCRIPTION

3.1 LONG SHORT-TERM MEMORY

LSTM is an advanced RNN, a sequential network that persists

information. It is able to handle the vanishing gradient problem faced
by RNNs. Recurrent Neural Networks, also called RNNs, are used for
persistent memory.

Suppose you remember a previous scene when watching a video, or

know what happened in the last chapter when reading a book. Also,
RNNs work in that they remember previous information and use it to
process the current input. The disadvantage of RNNs is that they
cannot remember long-term dependencies due to vanishing gradients.
LSTMs are explicitly designed to avoid long-term dependency issues.

At a high level, LSTMs work much like RNN cells. This is the inner
workings of LSTM networks. LSTM consists of three parts as shown
in the figure below, each performing a separate function.
20
 Figure 3.1: Layers of LSTM

The first part chooses whether the previous timestamp information

should be remembered or irrelevant and can be forgotten. In the
second part, the cell tries to learn new information from the input of
this cell. Finally, in the third part, the unit passes the updated
information from the current timestamp to the next timestamp.

These parts of an LSTM cell are known as gates. The first part is
known as Forget gate, the second part is called the Input gate and the
last one is named as the Output gate.

Figure 3.2: LSTM Functioning

21
3.2 MIN MAX SCALING

The main idea behind normalization/standardization is always the

same. Variables measured at different scales do not contribute equally
to the fitting and learning functions of the model and may end up being
biased.

Therefore, to solve this potential problem, normalization by feature,

such as MinMax scaling, is usually used before fitting the model.
Mathematically this can be expressed as

Figure 3.3: Min Max Scaling Formula

22
 CHAPTER 4

SYSTEM ANALYSIS

4.1 SOFTWARE REQUIREMENTS

As we know we need to run Python code for our project, we can

proceed with installing all the libraries that we need to run this
project. We also have requirements for a browser with which we can
access the internet. Our code can run in any Python Integrated
Development Environment (IDE). But for our project we’ve chosen to
use Google Collab as it is convenient to use it. And any computer,
laptop having any operating system can run the code using Google
Collab.

23
4.2 HARDWARE REQUIREMENTS

There is no necessary cap on hardware power or usage limit. We can

use any pc or laptop with enough memory to run command prompt or
terminal in os. This program can run on devices which have 2gb
memory as basic, further we can run it more smoothly when provided
solid state drives which can be quick in processing data.We need
enough storage as our data gets accumulated every time we search
movies on our program.

CHAPTER 5

SYSTEM IMPLEMENTATION

In our project we have gathered data from Yahoo Finance. Our dataset
consists of historical data of the following stock

DATASET: Microsoft share price – historical data

24
 Figure 5.1: Dataset – Microsoft Historical data

CHAPTER 6

METHODOLOGY

The model was implemented on Jupiter Notebooks using

python libraries like pandas, matplotlib, NumPy and yahoo
finance. The objective was to predict stock prices using
numerical, fundamental and sentiment analysis of companies.

25
The data was first imported from the yahoo finance library of
share prices which records all the changes in the stock prices on
an interval of 5 minutes as provided by the share’s stock
exchange – NASDAQ, DOW JONES, NIFTY etc.

In this particular testing model, we have used the Microsoft

share and we have created the data frame based on the closing
prices of the share daily and plotted them using matplotlib. A
minmax scaler was then applied onto it for the purpose of
rescaling all the values in the scale [0,1] and the reshaped model
was then trained using Long Short-Term Memory and the
model was then successfully compiled.

Fig 6.1: Graph of the closing price of Microsoft plotted

To analyze the true value of a stock - two measures are taken -

Price to earnings ratio (P/E ratio) The price to earnings ratio -
also known as the price to earnings ratio or price to earnings
ratio - is a financial measure of a company's stock price relative
to its earnings per share. In short, it shows the balance between
stock prices and earnings. Thanks to this ratio, we can see how
profitable it is to buy shares of a particular company.

26
We can also use the P/E ratio to determine whether a stock is
overvalued or undervalued. For example, if you consider two
companies in the same industry but which have completely
different values for the P/E ratios, it may mean that one of them
is not credibly valued.

Fig 6.2: Basic EPS

The P/E ratio can be calculated based on basic EPS (earnings

per share) by dividing the current share price by earnings per
share. Another measure is the price-to-sales ratio, also known as
the ratio. Since sales are often referred to as earnings, the P/S
ratio is also often referred to as the price-to-earnings ratio or the
price-to-earnings ratio. It is mainly used to measure the current
market value of a particular stock.

This ratio is often used in conjunction with the well-known

price-earnings ratio to determine a company's attractiveness
relative to its peers. The lower the price-to-sales ratio, the more
likely the company is undervalued and the more eligible a stock
is to “buy”. For example, if Company Y's price-to-sales ratio is
1.5 times, Company Y may be considered undervalued relative
to Company X. Therefore, it is reasonable to buy Company Y
and to sell company X.

27
Although the price-to-sales ratio is a solid investment measure,
it is recommended that you only compare it to similar
companies. Comparing the P/E ratio of an oil and gas company
like Shell to that of a technology company like Google is
pointless because the two operate very differently.

Fig 6.3: P/S Ratio

Using the above metrics, the data was scaled with the help of a
MinMax Scaler and prepped for testing and the LSTM principles were
applied on the data. The Long-Term Short-Term Memory Network is
an advanced form of Recurrent Neural Network , a sequential network,
that allows information to persist. It can handle the vanishing gradient
problem faced by RNNs. A cyclic neural network is also known as
RNN and is used for persistent memory. Note for the LSTM
layer, units is the number of LSTM neurons in the layer. 50 neurons
will give the model high dimensionality, enough to capture the
upwards and downward trends. return_sequences is True as we need to
add another LSTM layer after the current
one. input_shape corresponds to the number of time stamps and the
number of indicators. For dropout, 20% of 50 neurons will be ignored
randomly during each iteration of training.

28
Following the above same method, added 2nd, 3rd, and 4th LSTM
layer. The data was then scaled back to the initial format and the
model’s predictions were compared with the actual closing price. 

Fig 6.: LSTM Implementation

CHAPTER 7

CONCLUSION

29
The trained model was able to predict closing prices for the
dataset to within 0.93 points of the closing price and did well in
the long-term predictions. The main area where the prediction
was off by a margin was the Covid crash of 2020 which could
not possibly have been predicted by company values and
historical data. However, to solve this issue – sentiment analysis
can be used on news headlines pertaining to the stock in
question and call/put options can be provided accordingly.

CHAPTER 8

30
 REFERENCES

[1] Chiu, D. Y., & Chen, P. J. (2009). Dynamically exploring the

internal mechanism of the stock market by fuzzy-based support vector
machines with high dimension input space and genetic algorithms.
Expert Systems with Applications, 36(2), 1240-1248.

[2] Chong, E., Han, C., & Park, F. C. (2017). Deep learning
networks for stock market analysis and prediction: Methodology, data
representations, and case studies. Expert Systems with Applications,
83, 187-205.

[3] Dai, W., Wu, J. Y., & Lu, C. J. (2012). Combining nonlinear
independent component analysis and neural network to predict Asian
stock market indexes. Expert systems with applications, 39(4), 4444-
4452.

[4] Das, S. P., & Padhy, S. (2012). Support vector machines for
prediction of futures prices in the Indian stock market. International
Journal of Computer Applications, 41(3).

[5] Dash, R., & Dash, P. K. (2016). A hybrid stock trading

framework integrating technical analysis with machine learning
techniques. The Journal of Finance and Data Science, 2(1), 42-57.

31
[6] Enke, D., & Thawornwong, S. (2005). The use of data mining
and neural networks for forecasting stock market returns. Expert
Systems with applications, 29(4), 927- 940.

[7] Goodhue, D. L., & Thompson, R. L. (1995). Task-technology fit

and individual performance. MIS quarterly, 213-236.

[8] Guresen, E., Kayakutlu, G., & Daim, T. U. (2011). Using

artificial neural network models in stock market index prediction.
Expert Systems with Applications, 38(8), 10389-10397.

[9] Holland, J. H. (1992). Adaptation in natural and artificial

systems: an introductory analysis with applications to biology, control,
and artificial intelligence. MIT press.

[10] Jasic, T., & Wood, D. (2004). The profitability of daily stock
market indices trades based on neural network predictions: Case study
for the S&P 500, the DAX, the TOPIX and the FTSE in the period
1965–1999. Applied Financial Economics, 14(4), 285-297.

[11] Kim, H. J., & Shin, K. S. (2007). A hybrid approach based on

neural networks and genetic algorithms for detecting temporal patterns
in stock markets. Applied Soft Computing, 7(2), 569-576.

[12] Kim, K. J., & Han, I. (2000). Genetic algorithms approach to

feature discretization in artificial neural networks for the prediction of
stock price index. Expert systems with Applications, 19(2), 125-132.

32
[13] Kim, K. J., & Lee, W. B. (2004). Stock market prediction using
artificial neural networks with optimal feature transformation. Neural
computing & applications, 13(3), 255-260.

APPENDIX A

IMPLEMENTATION SCREENSHOTS
33
34
35
36
37
38