Master Project

MASTER PROJECTon

Prediction of the Stock Market (Web Application)Submitted in

Partial fulfillment of

Master of Business Administration (IT)

(2014-2016)

Internal guide: External Guide:Dr. Shailendra Kumar Dr.Amrita ChaturvediAssistant Professor Assistant ProfessorIIIT-Allahabad IIIT-Allahabad

Submitted By:Thirupathi JadiIMB2014002

Introduction of Stock MarketThe stock market is the market in which shares of publicly held companies are issued and traded

either through exchanges or over-the-counter markets.

Stock market prediction is the act of trying to determine the future value of a company stock or other financial instrument traded on an exchange. The successful prediction of a stock's future price could yield significant profit.

Problem Definition

Why do we need to predict the stock market? There are two prices that are critical for any investor to know, The current price of the investment

he or she owns or plans to own and its future selling price.Despite this, Investors are constantly reviewing past pricing history and using it to influence their

future investment decisions.Some investors won't buy a stock or index that has risen too sharply Because they assume that

it’s due for a correction, While other investors avoid a falling stock, because they fear that it will continue to deteriorate.

Diagnosis/MethodologyBy the help of an internet I searched many websites and Research papers then I found many

software’s are available for stock market prediction but those software’s difficult to understand and usage for non-finance professionals. Those software’s are not free.

I had spoken to few investors face to face in my local area. they told me that they are not able to understand financial terms but they are investing in stock market.

Then I collected 5 years historical data of National stock exchange, this data includes open, close, high, low, volume, adjacent close stock market attribute.

I had installed Hadoop, Mahout, Java, Maven, Rstudios..I have analyzed this data by the help of Machine Learning Techniques in Hadoop and RStudios,

Mahout.

Result

Historic Data of NSE

Processing of Data

LITERATURE REVIEWResearch Paper1.Scalable System for Textual Analysis of Stock Market Prediction:In this paper, after the preliminary result, they proved that Jieba can scale well with their file

parallelism version, i.e., scaling-up with four cores gets 80% faster compared to one core environment and scaling-out makes the linear improvement. The main factors of scalability are the nature of data independence and data replication. Text segmentation process can be executed in a parallel way. However, the performance improvement of Mahout is limited by the first step, i.e., file sequential process. Because the bottleneck has a huge impact on an overall response time of classification model training process, we have to deal with it in the future. They are still solving this problem by tracing Mahout source code. It is worth mentioning that Mahout starts to move its focus on Spark, a new popular large scale data processing project stating faster processing speed because of in-memory computation. They will use Spark to solve the scalability issue of machine learning function in another way.

Websites2.National Stock Exchange (NSE) website:The National Stock Exchange (NSE) is India's leading stock exchange covering various cities and

towns across the country. NSE was set up by leading institutions to provide a modern, fully automated screen-based trading system with national reach. The Exchange has brought about unparalleled transparency, speed & efficiency, safety and market integrity. It has set up facilities that serve as a model for the securities industry in terms of systems, practices, and procedures.

3.NiftysureshotThis site is focusing on intraday purpose. Their research team of analysts focus on very few types of

scrip i.e. Nifty future tips, stock option tips all are on an intraday basis. 4.InvestopediaIn this website, they are providing theoretical knowledge. They are not giving future tips.5.SharekhanHere they provide lots of services and information but those services are not free. Selection of data

or services is difficult for non-financial professionals.

LimitationsWe have to handle large scale of data.Many more financial calculations.Software’s are paid versions.Critical to Understand.Lack of Speed.

Software Requirement & Specification

Hadoop Mahout Java, RstudiosArtificial neural networks Predictive Algorithm Regression Technique Historical Data of NSE Stock Market

Scope of future work I have done stringent analysis of the state of the art research in the area of stock market analysis

and prediction. I have also found the limitations of the existing works and am planning to build on that.

Now I will design the web application along with solutions. By this web application Investors easy to understand stock market movements in the future. This application is easy to use of non-financial professionals. This application is plug and play model. This application is open source for investors.

References1.Roy Guanyu Lin and Tzu-Chieh Tsai,They have proposed a scalable system for textual analysis of the market prediction, at DATA ANALYTICS 2014:the third international conference on data analytics.2.The national stock exchange website www.nseindia.com3. www.niftysureshot.com4.www.investopedia.com5.www.sharekhan.com

THANK YOU