Stock Market Prediction Using Artificial Neural Network

International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-9, Sept- 2016]
Infogain Publication (Infogainpublication.com) ISSN : 2454-1311
www.ijaems.com Page | 1416
Stock Market Prediction Using Artificial Neural
Network
Nazish Nazir1
, Mudasirahma Dmutto2
1
RN Engineering & Management College Rohtak, Haryana, India
2
Department of CSE, Assistant Professor, SSM College of Engineering &Technology, Jammu and Kashmir, India
Abstract— Nowadays during increasingly developed
technology of the World Wide Web and Internet, the data
is becoming extremely rich. With the application of data
recognition process, the information extracted from data
has become the most important part in some areas of
society, management field, finance and markets, etc. It is
necessary to develop the valid method to understand the
knowledge of the data. Whether you are looking for good
investments or are into stock trading, stock prediction or
forecast plays the most crucial role in determining where
to put in the money or which stock to be acquired or sold.
Keywords— Artificial Neural Network, Stock Market,
Forecast Algorithms.
I. INTRODUCTION
An investment theory suggests what parameters one
should take into account before placing his (or her) capital
on the market. Traditionally the investment community
accepts two major theories: the Firm Foundation and the
Castles in the Air. Reference to these theories allows us to
understand how the market is shaped, or in other words
how the investors think and react. It is this sequence of
‘thought and reaction’ by the investors that defines the
capital allocation and thus the level of the market. There
is no doubt that the majority of the people related to stock
markets is trying to achieve profit. Profit comes by
investing in stocks that have a good future (short or long
term future). Thus what they are trying to accomplish one
way or the other is to predict the future of the market. But
what determines this future? The way that people invest
their money is the answer; and people invest money based
on the information they hold.
The factors that are under discussion on this schema are:
the content of the ‘Information’ component and the way
that the ‘Investor’ reacts when having this info.
According to the Firm Foundation theory the market is
defined from the reaction of the investors, which is
triggered by information that is related with the ‘real
value’ of firms.
Whether you are looking for good investments or are into
stock trading, stock prediction or forecast plays the most
crucial role in determining where to put in the money or
which stock to be acquired or sold. Market trends often
reflect the mood of the market and not essentially the
status of a company or the true value of the stocks. It is
often that stock prices soar based on external factors and
it is not uncommon to find stock traders and investors to
base their decisions on current affairs and market trends
while trying to forecast the stock of any specific
company.
The two stock forecasting methods any investor or stock
trader must use are the fundamental Research and Stock
Forecast Algorithms.
Fundamental Research is a mandatory method for any
investor. The method involves meticulous studying of a
company’s financial health, the value of assets, debts,
cash, revenues, expenses, profitability and plans of
development. Fundamental Research is a well rounded
stock prediction method for all the data that actually
matters are taken into consideration while determining the
true value of a stock
Stock Forecast Algorithms are aimed at making the best
use of the right time, right price and the right quantity of
stocks that must be traded. The Algorithm in place helps a
trader to forecast the time at which the price would be the
most favourable to either buy or sell a stock. The system
predicts absolutely on numbers and has not even remotely
affected by popular emotions. Finally, one should not get
caught up in the daily trading, and miss out on global
trends.
There are several methods that have been developed for
the purpose of Stock Prediction Methods such as
Fundamental Analysis, Technical Analysis, Data Mining
Technology, Internet Based Data Sources of Stock Market
Prediction and Application of Complexity Science for
Stock Prediction etc. These methods are: Fundamental
Analysis, Technical Analysis, Data Mine Technologies
etc.
II. INDICATORS FOR STOCK PREDICTION
There are several indicators that are used for the purpose
of stock prediction like:
Moving average: The moving average calculates the
average of past n values till today. A moving average is
commonly used with time series data to smooth short

term noisy data and highlight longer-term trends or
cycles.
Simple moving average: A simple moving average
(SMA) is the un-weighted mean of the previous data; the
data can be in n number which is used in financial
applications.
Exponential moving average:In EMA we will give more
weight age to recent values than historical data. For
example we will give more weight age to data of 2015
rather than data of 2014.
Rate of change: The ratio of current price to price n
quotes earlierN is generally 5 to 10 days. For example we
will check here recent data and ups and downs in recent
days rather than older days…
Relative Strength Index: Measure the relative size of
recent upward trends against the size of downward trends
within specified time interval. It will check ups in market
prices and downs in market prices.
Process of Stock Prediction:
III. RELATED WORK
Raluca-Mariana tefan [2012] states that there are
different methods of data classification. In it we divide
data into categories for use them at highest level. Which
will provide effectiveness and efficient data?
Ian H. Witten[2013] determines that Text mining is a
multidisciplinary field which is involved in information
retrieval, text analysis, information extraction, clustering,
categorization, database technology, machine learning
and data mining.
MAHESH T R et al.[2010] explain thatText mining is a
new research area in which we mine knowledge from
unstructured text. It means we mine needed text /
information using text mining.
Simon Hawkinset al. [2014] concluded that extract
information from different written resources. Thus text
mining method can be used on various fields like social
media, news etc. Outlier detection is used to remove
fluctuations from data.
Andrea Nemeti a et al. [2015] determine that Stock
market prediction is a vast area of research where we can
apply text mining techniques for retrieve the text and also
use SVM for retrieve numerical data sets and text data.
After prediction of stock prices we get collection of data
sets which is unstructured.
Alberto Arteta Albert1 et al. [2015] gave a conclusion
thatusing prediction we can predict future stock prices
that they will increase or decrease. If successful
prediction of stock then it will be beneficial for future and
with stock market prediction profit will also increase.
There are moving averages algorithms like SMA,EMA,
ROI will be use for stock market prediction.
NilamUpasaniaet al. [2015] explains that we have to
make unstructured data into structured data and remove
noisy data from data sets. Noisy data can be missing
value, duplicated data, etc. Removing of noisy data is also
called outlier detection which use different mining
methods for detection of noise and then remove it.
Forecasting stock market movement direction with
support vector machine
IV. OBJECTIVES AND METHODOLOGY OF
THE WORK
The objective of the work that is performed involves
collection and pre processing the stock market data and
applying filters that are moving average, weighted,
triangular, exponential on the pre-processed stock market
data and further applying neural network to predict
performance of stock and to compare the performance of
proposed and existing system.
Research methodology:

International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol
Infogain Publication (Infogainpublication.com
www.ijaems.com
Facilities used
My proposed work plan for research is as following:
Data Collection: - Firstly data will collect from
companies, news, and social media.
Training data: - All training data of stock market will save
in excel sheet using MS_EXCEL.
Test data: Test data will from companies and store test
data of stock market in excel sheet.
Simulator:- MATLAB r2010a will be useas a simulator
for stock market prediction.
After prediction of stock price our objective is to remove
fluctuations from data using moving average algorithms in
MATLAB
Results:
The figure below shows Results if filters applied on the
stock data set
International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol
Infogainpublication.com)
My proposed work plan for research is as following:-
Firstly data will collect from
All training data of stock market will save
Test data will from companies and store test
MATLAB r2010a will be useas a simulator
prediction of stock price our objective is to remove
fluctuations from data using moving average algorithms in
The figure below shows Results if filters applied on the
The figure below shows the functionality of neural
network on the stock data set
The figure below shows the validation performance of
dataset using neural network
V. CONCLUSION
The extracted data can be used for applications ranging
from market analysis. It decomposes the original stock
indexinto the trend terms, the market fluctuation terms,
the noise terms and time series with different
economicfeatures. Then SVM will use all moving average
algorithm to predict stock prices. After prediction of stock
prices using all algorithms the result of all
combine. Then moving average algorithms result will
check the duplicate and missing values in data and
) ISSN : 2454-1311
Page | 1418
The figure below shows the functionality of neural
twork on the stock data set
The figure below shows the validation performance of
CONCLUSION
The extracted data can be used for applications ranging
from market analysis. It decomposes the original stock
the trend terms, the market fluctuation terms,
the noise terms and time series with different
economicfeatures. Then SVM will use all moving average
algorithm to predict stock prices. After prediction of stock
prices using all algorithms the result of all algorithm will
combine. Then moving average algorithms result will
check the duplicate and missing values in data and

remove noisy data. In future we can integrate fuzzy logic
in it to smooth out the data and remove the fluctuations.
Thus after prediction we remove the noisy data and also
fill the missing values. So the result of prediction will be
clean data without any noisy and missing values.
REFERENCES
[1] Uğuz and Harun, “A two-stage feature selection
method for text categorization by using information
gain, principal component analysis and genetic
algorithm”, Knowledge-Based Systems, vol. 24, no.
7, (2011), pp. 1024-1032.
[2] Beneki,C.,Eeckels,B.,andLeon,L.,Signal Extraction
and Forecasting of the UK Tourism Income Time
Series: A Singular SpectrumAnalysis Approach.
Journal of Forecasting; 2012, 31(5), p. 351-400.
[3] NeelamadhabPadhy , Dr. Pragnyaban Mishra and
andRasmitaPanigrahi (2012) “The Survey of Data
Mining Applications And Feature
Scope”International Journal of Computer Science,
Engineering and Information Technology
(IJCSEIT), Vol.2, No.3, June 2012
[4] Zhang Xuegong. About the statistical learning
theory and support vector
machine.ACTAAUTOMATICA SINICA; 2000, 01, p.
36-46.
[5] Ajay Kumar Pal ,Saurabh Pal “Evaluation of
Teacher’s Performance: A Data Mining Approach
“International Journal of Computer Science and
Mobile Computing ,Vol. 2, Issue. 12, December
2013, pg.359 – 369.
[6] 2nd International Conference on Information
Technology and Quantitative Management,ITQM
2014Stock Price Prediction Based on SSA and SVM
by WEN Fenghuaa*, [6] XIAO Jihongb, HE
Zhifanga, GONG XuaMa. Business School of
Central South University, Changsha, 410081,
Chinab. School of Economics and Management,
Changsha University of Scienceand Technology,
Changsha, 410004 ,China
[7] W. Huang et al., “Forecasting stock market
movement direction with support vector machine,”
Computers & Operations Research, 32, pp., 2005
[8] J. Moody, et al., “Learning to trade via direct
reinforcement,” IEEE,Transactions on Neural
Networks, vol. 12, no. 4, Jul. 2001.
[9] S. Zemke, “On developing a financial prediction
system: Pitfall and possibilities,” Proceedings
of DMLL-2002 Workshop, ICML, Sydney,
Australia, 2002.

Stock Market Prediction Using Artificial Neural Network

More Related Content

What's hot (20)

Similar to Stock Market Prediction Using Artificial Neural Network (20)

Recently uploaded (20)

Stock Market Prediction Using Artificial Neural Network