Stock Market Prediction Using Machine Learning

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328930285

Stock Market Prediction Using Machine Learning

Research Proposal · November 2018


DOI: 10.13140/RG.2.2.12300.77448

CITATIONS READS

0 5,636

1 author:

Kranthi Sai Reddy Vanukuru


Sreenidhi Institute of Science & Technology
4 PUBLICATIONS   0 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Stock Market Prediction Using Machine Learning View project

Intelligent Smart Zone Based Vehicle Speed Control System Using RF View project

All content following this page was uploaded by Kranthi Sai Reddy Vanukuru on 16 November 2018.

The user has requested enhancement of the downloaded file.


International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 10 | Oct 2018 www.irjet.net p-ISSN: 2395-0072

Stock Market Prediction Using Machine Learning


V Kranthi Sai Reddy1
1Student, ECM, Sreenidhi Institute of Science and Technology, Hyderabad, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - In the finance world stock trading is one of the highly fluctuating and it effects the investor’s belief. Stock
most important activities. Stock market prediction is an act of prices are considered to be a very dynamic and susceptible
trying to determine the future value of a stock other financial to quick changes because of underlying nature of the
instrument traded on a financial exchange. This paper financial domain and in part because of the mix of a known
explains the prediction of a stock using Machine Learning. The parameters (Previous day’s closing price, P/E ratio etc.) and
technical and fundamental or the time series analysis is used the unknown factors (like Election Results, Rumors etc.).
by the most of the stockbrokers while making the stock There has been numerous attempts to predict stock price
predictions. The programming language is used to predict the with Machine Learning. The focus of each research projects
stock market using machine learning is Python. In this paper varies a lot in three ways. (1) The targeting price change can
we propose a Machine Learning (ML) approach that will be be near-term (less than a minute), short-term (tomorrow to
trained from the available stocks data and gain intelligence a few days later), and a long-term (months later), (2) The set
and then uses the acquired knowledge for an accurate of stocks can be in limited to less than 10 particular stock, to
prediction. In this context this study uses a machine learning stocks in particular industry, to generally all stocks. (3) The
technique called Support Vector Machine (SVM) to predict predictors used can range from a global news and economy
stock prices for the large and small capitalizations and in the trend, to particular characteristics of the company, to purely
three different markets, employing prices with both daily and time series data of the stock price.
up-to-the-minute frequencies.
The probable stock market prediction target can be the
Key Words: Stock Market, Machine Learning, Predictions, future stock price or the volatility of the prices or market
Support Vector Machine trend. In the prediction there are two types like dummy and
a real time prediction which is used in stock market
1. INTRODUCTION prediction system. In Dummy prediction they have define
some set of rules and predict the future price of shares by
Basically, quantitative traders with a lot of money from calculating the average price. In the real time prediction
stock markets buy stocks derivatives and equities at a cheap compulsory used internet and saw current price of shares of
price and later on selling them at high price. The trend in a the company.
stock market prediction is not a new thing and yet this issue
is kept being discussed by various organizations. There are Computational advances have led to introduction of
two types to analyze stocks which investors perform before machine learning techniques for the predictive systems in
investing in a stock, first is the fundamental analysis, in this financial markets. In this paper we are using a Machine
analysis investors look at the intrinsic value of stocks, and Learning technique i.e., Support Vector Machine (SVM) in
performance of the industry, economy, political climate etc. order to predict the stock market and we are using Python
to decide that whether to invest or not. On the other hand, language for programming.
the technical analysis it is an evolution of stocks by the
2. Methodology
means of studying the statistics generated by market
activity, such as past prices and volumes.
In this project the prediction of stock market is done by
In the recent years, increasing prominence of machine the Support Vector Machine (SVM) and Radial Basis Function
learning in various industries have enlightened many traders (RBF).
to apply machine learning techniques to the field, and some
2.1 Support Vector Machine
of them have produced quite promising results.
A Support Vector Machine (SVM) is a discriminative
This paper will develop a financial data predictor
classifier that formally defined by the separating hyperplane.
program in which there will be a dataset storing all historical
In other words, the given labeled training data (supervised
stock prices and data will be treated as training sets for the
learning), the algorithm outputs the optimal hyperplane
program. The main purpose of the prediction is to reduce
which categorizes new examples. In the two-dimensional
uncertainty associated to investment decision making.
space this hyperplane is a line dividing a plane into two parts
Stock Market follows the random walk, which implies where in each class lay in either side.
that the best prediction you can have about tomorrow’s
Support Vector Machine (SVM) is considered to be as
value is today’s value. Indisputably, the forecasting stock
one of the most suitable algorithms available for the time
indices is very difficult because of the market volatility that
series prediction. The supervised algorithm can be used in
needs accurate forecast model. The stock market indices are

© 2018, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1032
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 10 | Oct 2018 www.irjet.net p-ISSN: 2395-0072

both, regression and classification. The SVM involves in a center, so that. Any function which satisfies the property is
plotting of data as point in the space of n dimensions. a radial function.

These dimensions are the attributes that are plotted on RBF = Local Response Function
particular co-ordinates. SVM algorithm draws a boundary
over the data set called as the hyper-plane, which separates The RBF Kernel is nothing more than a low-band pass
the data into two classes as shown in the Fig 1. filter, which is well known in Signal Processing as a tool to
smooth images. RBF Kernel acts as the prior that selects out
smooth solutions.

The Radial basis function kernel, is also called as the RBF


kernel, or Gaussian kernel, is a kernel that is in the form of a
radial basis function (more specifically, a Gaussian function).
The RBF kernel is defined as

KRBF (x, x’) = exp [-y ||x -x’||2]

Where y is the parameter that sets “spread” of the kernel

Fig -1: The Support Vector Machine Decision Making


Boundary

The hyper-plane is a decision boundary which is later


extended or maximized on either side between the data
points. Considering the same figure, if µ is some unknown
data point and w is vector which is perpendicular to the
hyper-plane, then the SVM decision rule will be

………………………………………………………… (1) Fig -2: RBF Network


The width w of the hyper-plane must be maximized the The RBF units provide a new basis set for synthesizing the
spread output function. The radial basis functions are not
orthogonal and are overcomplete.
w = [2/ || w ||] …………………………………………………. (2)
2.3 The Learning Environment
w = (max [2/|| w ||]) ………………………………………... (3)
The Weka and the YALE Data Mining Environments were
Applying lagrange’s multiplier as used for carrying out the experiments. The general setup
used is as follows:
L = 0.5|| w ||2 -∑ ……………. (4)

L=∑ …………………………… (5)

The updated decision rule will be

(∑ )µ+b ≥ 0

2.2 Radial Basis Function (RBF)

In the machine learning, the radial basis function kernel,


or RBF kernel, is a popular kernel function used in the
various kernelized learning algorithms. In particular, it is
most commonly used in support vector machine
classification.

A radial basis function is the real-valued function whose


value depends only on the distance from the origin, so that;
or alternatively on the distance from some other point, called Fig -3: Learning Environment

© 2018, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1033
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 10 | Oct 2018 www.irjet.net p-ISSN: 2395-0072

3. Model Creation and Evaluation Methods Step 2: In the next step the data value of any stock that can
be converted into the CSV file (Comma Separate Value) so
In this paper we focus on predicting the Stock Market that it will easily load into the algorithm.
using Machine Learning model i.e., Support Vector Machine
(SVM) by RBF kernel. Step 3: In the next step in which GUI is open and when we
click on the SVM button it will show the window from which
3.1 Feature Selection we select the stock dataset value file.
In this project we use four features to predict stock price
Step 4: After selecting the stock dataset file from the folder it
direction – price volatility, price momentum, sector
will show graph Stock before mapping and stock after
volatility, and sector momentum. More details are provided
mapping.
in Table 1, styled in the form used by Kim [4].

Table 1: Features used in SVM Step 5: The next step algorithm calculated the log2c and
log2g value for minimizing error. So, it will predict the graph
for the dataset value efficiently.
Feature Description Formula
Name
Step 6: In final step algorithm display the predicted value
σs Stock price volatility. graph of select stock which shows the original value and
This is an average predicted value of the stock.
over the past n days
of percent change in 4. RESULTS
the given stock’s
price per day.
Stock This is an average of
Momentum the given stock’s
momentum over the
past n days. Each day
is labeled 1 if closing
price that day is
higher than the day
before, and −1 if the
price is lower than
the day before.
σi Index volatility. This
is an average over
the past n days of
percent change in the
index’s price per day.
Index This is an average of
Momentum the index’s Fig -4: Stock dataset for IBM Inc. in CSV file
momentum over the
past n days. Each day
it is labeled 1 if
closing price that day
is higher than the day
before, and −1 if the
price is lower than
the day before

3.2 Steps for Stock Market Prediction

Step 1: This step is important for the download data from the
net. We are predicting the financial market value of any
stock. So that the share value up to the closing date are
download from the site.

Fig -5: Predicted Output by SVM for IBM Inc.

© 2018, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1034
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 10 | Oct 2018 www.irjet.net p-ISSN: 2395-0072

5. CONCLUSION

In the project, we proposed the use of the data collected


from different global financial markets with machine
learning algorithms in order to predict the stock index
movements. SVM algorithm works on the large dataset value
which is collected from different global financial markets.
Also, SVM does not give a problem of over fitting. Various
machine learning based models are proposed for predicting
the daily trend of Market stocks. Numerical results suggest
the high efficiency. The practical trading models built upon
our well-trained predictor. The model generates higher
profit compared to the selected benchmarks.

REFERENCES

[1] Zhen Hu, Jibe Zhu, and Ken Tse “Stocks Market
Prediction Using Support Vector Machine”, 6th
International Conference on Information Management,
Innovation Management and Industrial Engineering,
2013.M.
[2] Wei Huang, Yoshiteru Nakamori, Shou-Yang Wang,
“Forecasting stock market movement direction with
support vector machine”, Computers & Operations
Research, Volume 32, Issue 10, October 2005, Pages
2513–2522.
[3] N. Ancona, Classification Properties of Support Vector
Machines for Regression, Technical Report, RIIESI/CNR-
Nr. 02/99.
[4] K. jae Kim, “Financial time series forecasting using
support vector machines,” Neurocomputing, vol. 55,
2003.
[5] Debashish Das and Mohammad shorif uddin data mining
and neural network techniques in stock market
prediction: a methodological review, international
journal of artificial intelligence & applications, vol.4,
no.1, January 2013

© 2018, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1035

View publication stats

You might also like