Web Based Stock Forecasters - Winlab

1y ago
2.88 MB
82 Pages
Last View : 3d ago
Last Download : 3m ago
Upload by : Melina Bettis

1 Page2008SOFTWARE ENGINNERING OFWEB APPLICATIONSGROUP # 5WEB BASEDSTOCK FORECASTERSThis report describes the Stock Prediction System titled “TIMINGTHE MARKET” developed by our team which has various moduleslike Data Mining, Web Services, Neural Network Based StockPrediction and Technical Indicators.SUBMITTED BY:AMARINDER CHEEMAATEET VORACHETAN JAINPUNEET KATARIARONAK SHAHSIDDHARTH WAGH7 May, 2008

2 PageBreakdown of responsibilities:

3 PageTABLE OF CONTENTS:Breakdown of Responsibilities.01Summary of Changes.06Glossary of Terms .071. Introduction.101.1 Project Goals and Requirements.111.2 System Requirements.121.3 Statistical comparison of our product with current software’s in market.132. System Description.142.1 System Block Diagram.142.2 System Description and general Description.152.3 Use cases.162.3.1 Use case Casual Description.172.3.2 Functional Specification Requirement.182.3.3 Interaction Diagram of key use cases.222.4 Activity Diagrams.252.5 System Class Diagram.302.5.1 Class Diagram Description.322.5.2 Class Diagram Attributes and Description.333. Data Mining.353.1 Need for Data Mining.353.2 Data Mining – An Overview.363.3 Features of the Fetch Script.373.4 Current Price Fetch.373.5 DB Schema Design.384. Web services.394.1 Procedure for using Web Services.394.1.1 High Level view of Web Services.404.1.2 Consuming a Web Service.414.1.3 Benefits of using ASP.net for creating Web Services.424.2 Web Service inclusion in project.434.2.1 Detailed description of Web Services developed for the project.454.3 Understanding WSDL.464.4 Soap Request and response.495. Stock Prediction using Neural Networks.505.1 Neural Network Technology.505.1.1 Artificial Neural Network Approach.515.1.2 Neural Network Training and Testing.515.1.3 Neuron Models.525.2 Matlab(v7.1)Neural Network Toolbox.555.3 Basic working of our Model.575.4 Detailed Description /Program Algorithm.585.5 Performance Evaluation.646. Stock prediction using indicators.666.1 Relative Strength Index.666.2 Stochastic Oscillator.67

4 Page6.3 Moving Averages.676.4 Price Momentum Oscillator.686.5 Money Flow Index.686.6 Demarker Indicator.696.7 Williams Indicator.706.8 Commodity Channel Index.716.9 Aroon Indicator.727. User Interface.747.1 Website.75Accomplishments.78Shortcomings.80Future Work.81References.82

5 PageTABLE OF FIGURES:Figure 1: Comparison of software’s.13Figure 2: System Block Diagram. 14Figure 3: Use Case Diagram.16Figure 4: Use Case # 2 and #3 Interaction Diagram.22Figure 5: Use Case # 5 Interaction Diagram.23Figure 6: Use Case # 7 Interaction Diagram.24Figure 7: Use Case # 1 Activity Diagram.25Figure 8: Use Case # 2 Activity Diagram.26Figure 9: Use Case # 3 Activity Diagram.27Figure 10: Use Case # 5 Activity Diagram.28Figure 11: Use Case # 7 Activity Diagram.29Figure 12: System Class Diagram.30Figure 13: Class Diagram for Data Mining.36Figure 14: Database Schema.38Figure 15: Basic Web Service Description.41Figure 16: Class Diagram for Web Services.44Figure 17: Multi layer Feed Network Topology.51Figure 18: Basic Neuron Model.52Figure 19: Basic Functionality of Neural Network.53Figure 20: Log-Sigmoid Transfer Function.54Figure 21: Tan- Sigmoid Transfer Function.54Figure 22: Linear Transfer Function.54Figure 23: Basic Working of our Model.57Figure 24: Class Diagram for Stock Prediction.58Figure 25: Class Diagram for User Interface.74

6 PageSUMMARY OF CHANGES: The report now mentions the number of companies for which we have developed ourStock Prediction Model which is 20. (Not mentioned in previous report) The PHP script imports data every 15 minutes because certain indicators implementformula which requires current stock volume and price to give correct prediction result. The system block diagram has been modified and a brief description is provided formore clarity which was missing in earlier report. The use case diagram has been modified as per requirement. The report now includes UML – Class, Interaction and Activity Diagrams for eachmodule of our system. Web services Class Diagrams with explanation of classes is now a part of the reportwhich was missing earlier. Database snapshot has now been replaced with a database schema diagram. Web services earlier written in Java have now been modified to ASP.NET due tointerface issues. References to text and diagrams have been duly provided where ever required.

7 PageGLOSSARY OF TERMS:Artificial IntelligenceThe field of computer science dedicated to producing programs that attempt to mimic theprocesses of the human brain.utocorrelationThe correlation between the values of a time series and previous values of the same timeseries.Back-Propagation NetworkA feed forward multilayered neural network that is a commonly used neural networkparadigm.Bear MarketA securities market characterized thus based on declining prices.Bull MarketA securities market characterized thus on rising prices.Buy and HoldThe acquisition of a tradable for the long term rather than quick turnover.ChartsA display or picture of a security that plots price and/or volume (the number of shares sold).Confidence LevelThe degree of assurance that a specified failure rate is not exceeded.Data miningProcess of sorting through large amounts of data and picking out relevant information.EpochPresentation of the set of training (input and/or target) vectors to a network and thecalculation of new weights and biases. Note that training vectors can be presented one at atime or all together in a batch.Fundamental AnalysisThe analytical method by which only the sales, earnings and the value of a given tradable'sassets may be considered.Gantt chartsBar charts illustrating the start and finish dates of the terminal elements and summaryelements of a project

8 PageHidden layerLayer of a network that is not connected to the network output (for instance, the first layer of atwo-layer feed-forward network).Input layerLayer of neurons receiving inputs directly from outside the network.Learning ruleLearning rule in which weights and biases are adjusted by error-derivative (delta) vectorsback-propagated through the network. Back-propagation is commonly applied to feed forwardmultilayer networks. Sometimes this rule is called the generalized delta rule.Momentum IndicatorA market indicator utilizing price and volume statistics for predicting the strength or weaknessof a current market and any overbought or oversold conditions, and to note turning pointswithin the market.Moving AverageA mathematical procedure to smooth or eliminate the fluctuations in data and to assist indetermining when to buy and sell.Neural NetworkAn artificial intelligence program that is capable of learning through a training process of trialand error.NeuronBasic processing element of a neural network. Includes weights and bias, a summingjunction, and an output transfer function. Artificial neurons, such as those simulated andtrained with this toolbox, are abstractions of biological neurons.Output layerLayer whose output is passed to the world outside the network.Pattern recognitionTask performed by a network trained to respond when an input vector close to a learnedvector is presented. The network “recognizes” the input as one of the original target vectors.RegressionA mathematical way of stating the statistical linear relationship between one independent andone dependent variableRelative Strength Index An indicator invented by J. Welles Wilder and used to ascertainoverbought/oversold and divergent situations.

9 PageResilient back-propagationTraining algorithm that eliminates the harmful effect of having a small slope at the extremeends of the sigmoid squashing transfer functions.Resistance LineOn a chart, a line drawn indicating the price level at which rising prices have stopped risingand have moved sideways or reversed direction.SkewA descriptive measure of lopsidedness in a distribution.Stochastic OscillatorAn overbought/oversold indicator that compares today's price to a preset window of high andlow prices. These data are then transformed into a range between zero and 100 and thensmoothed.Support LineOn a chart, a line drawn, indicating the price level at which falling prices have stopped fallingand have moved sideways or reversed direction.Technical AnalysisA form of market analysis that studies demand and supply for securities and commoditiesbased on trading volume and price studies using charts and modeling techniques.TickThe minimum fluctuation of a trading instrument.TrainingProcedure whereby a network is adjusted to do a particular job. Commonly viewed as anoffline job, as opposed to an adjustment made during each time interval, as is done inadaptive training.TrendThe general drift, tendency or bent of a set of statistical data as related to time.VectorColumn vector of weights coming from a neuron or input.VolumeThe shares which are traded for a given market or tradable within a specified time period.Web serviceSoftware system designed to support interoperable Machine to Machine interaction over anetwork.

10 P a g e1. INTRODUCTIONStock market prediction is the act of trying to determine the future value of a companystock or other financial instrument traded on a financial exchange. The successful prediction of astock's future price could yield significant profit. The stock market is not an efficient market.Herding behavior is common among investors, all investors do not get all information at thesame time and the time it takes to evaluate information before they act differs between investors.Many investors do not show rational behavior. Greed and fear are strong feelings and may resultin panic sales and stock market bubbles. Hence, to regulate the stock market to obtainmaximum profit or achieve a certain objective in general without falling prey to inconsistencies,predicting stock behavior is a pressing requirement.Prediction methodologies fall into two broad categories: fundamental analysis andtechnical analysis. Fundamental analysis of a business involves analyzing its income statement,financial statements and health, its management and competitive advantages, and itscompetitors and markets. It is more subjective compared to technical analysis. Fundamentalanalysis maintains that markets may misprice a security in the short run but that the "correct"price will eventually be reached. Profits can be made by trading the mispriced security and thenwaiting for the market to recognize its "mistake" and recalculate the security price. On the otherhand, Technical Analysis is an approach that uses information of past stock behavior in order toforecast future price movements. Within the technical analysis community there exist severalschools with different techniques, but they all have in common that they use price and volumehistory. A basic thought is that it takes time before the market reacts upon new information andthat pattern often occur in price behavior which makes forecasting possible.There are several factors that explain why technical analysis works:1. Most speculators on the market act upon fundamental analysis, so that kind of factsinfluence stock prices strongly. But all operators do not get this information at the sametime. When there are positive news of a company, those acting immediately can buyshares for a lower price than those getting the news later.2. Large investors such as mutual funds and banks are often not placing their whole blockorders at the same time when they are buying larger quantities of securities because thiswould risk triggering an unnecessary high price advance. Instead, the orders are spreadover a period that can last several weeks. The resulting increased purchase pressuremay result in a steady advancing trend under the period the purchases continue.3. It is more psychological stressing to go against the trend than to follow it. People areherding animals and like to do as others are doing. This is why a rising stock price is asignal in itself that the price will advance even more. Of course one has to be careful withstocks that have been rocketing, because they will often recoil.

11 P a g eHence, we limit our focus to technical analysis which has time and again proved its supremacyover other methods. There are many tools available to investors using technical analysis butnone of them removes entirely the element of chance from investment decisions. Large tradingorganizations can employ sophisticated computer systems and armies of analysts. We, asstudents, attempt to employ a simple set of formidable techniques to achieve the same result forthe benefit of small-time investors who cannot afford to hire experts or buy costly softwares tomake their investment decisions.1.1Project Goals and Requirements: Selection of target customers: Our customers are small-time daily or weekly investorswho trade on an individual basis and who do not have the time or resources to avail ofcommercial forecasting services or hire agents. Selection of information to be tracked: We aim to collect and use 10 indicators orpatterns for performing technical analysis and provide predictions. We also plan toprovide services like RSS feeds, current stock quotes, price charts, recommendationsand alerts to help our customer in making a wise investment decision. Data collection: Our major data source is the Google Finance service which provides uswith daily stock prices for an entire year. We mine current prices of stocks from YahooFinance. Charting: Based on the information in the database, we plan to display the stock pricesto the end-user using charts. Machine learning: The logic to recognize trends in market for past one year and makewise decisions based on statistical inference should be coded in the machine in the formof efficient algorithms. We plan to use technical indicators and neural networks to framesuch algorithms. Implement Web services: We aim to design web services to connect to predictionmodels that track different stocks as queried by user and issue forecasts about pricemovement of a given stock. Design web interface: A user-friendly web interface needs to be created and hosted toaid users to get valuable information and timely recommendations and tips about dealingwith their stock options.

12 P a g e1.2System requirements: The system should collect data from web sources like Yahoo Finance, Google finance ata time interval of 15 minutes and update its database from a set of 25 yahoo.com The system should maintain a host of web services that fairly link different servicemodules with client interface. Different prediction models need to be built that give a fair recommendation to thecustomer whether to buy/sell/hold stock. The system should train itself from a set of pastdata and simulate on a remaining set of test data to improve its precision. The system should aggregate all indications given by technical indicators and try toprovide a recommendation that supports the more robust neural network system. The system should track various stock movements simultaneously that might not beowned currently by customers and send alerts if it notices a favorable tilt in the graph thathints at immediate purchase. A friendly user interface should be designed for customers and access to specificservices should be restricted to only registered users to dissuade hackers fromdestabilizing the system. Various customer profiles should be maintained by the system and confidential customerdata should be protected at all costs from prying softwares. The system should regularly send emails or short messages to the customer informinghim about the current trends and future scenarios. It could feed the user current news andconduct surveys to judge its performance. The system may provide in-detail risk analysis of user portfolios to generate Sharpe Ratioto evaluate their current and future risk standing in the market and take appropriatecorrective measures.

13 P a g e1.3 Statistical comparison of our product with current software’s in t-inindicatorsData feedOnline/DownloadAlertsOptimal TraderYesNo15EOD,DelayedDNoNinja yedDNoTA- LibNoNo125NoDNoMeta OLNo“TimingMarket”theFig. 1: Comparison of software’s (Courtesy: Wikipedia – Technical Analysis)Terminology:Charting: Drawing elaborate charts based on stock information stored in database.Verification: Providing recommendations supporting the fundamental predictionBuilt-in indicators: Various indicators used for studying trends like Dow, S&P 500 etc.Data feed: How the data is extracted from web sourcesi.EOD: End of dayii.RT: Real-timeiii.Delayed: After a specific time intervalOnline/Download: Whether the software is available online or is to be downloaded.Alerts: Email, web, SMS services.

14 P a g e2. SYSTEM DESCRIPTION2.1 System Block Diagram:Fig 2: System Block Diagram

15 P a g e2.2 System description and general working:As shown in Fig: 3, the system under consideration has three levels: presentation level, serviceslevel and storage level. The user gives the input and gets the required output through theinterface. The services level performs various actions on the input data according to the user’sdemands and send the data back to the user. This level consists of Neural Network, WebServices and the indicators. The storage level lies in the bottom of the other levels and has thedatabase that store and records the entire data.The user is provided with the “home page” when he opens the application. The user is asked toprovide a correct combination of username and password. This pair of username and passwordis sent to the database for verification. If the combination is correct, the user is directed to the“service page”. If the user is new to the system then he is allowed to register as a new user. Theuser selects the “registration” button and is then allowed to fill in the registration form, when thisis done, the corresponding information is sent and stored into the database and the user hasnow got the username and password to login into the system.There are three key options that a user can select according to his requirement. These are: Get Quote (to get the quotes for a selected stock) Get Graphs (to get the graphs for a selected stock) Get Prediction (to get the prediction for a selected stock).Once the user enters the home page, he can select a stock from the drop down list and thenselect one of these options accordingly.If a user wants to see the graph of a particular stock, the information is directly sent to thedatabase, corresponding data is fetched, required actions are performed on it and the requiredgraph is sent back to the top layer that is the interface. If the user wants to get the quotes of aparticular stock, he selects that stock and clicks the “quotes” button. In this case the WebService is called which takes the data from the data base, performs the required actions andsend the required data to the user.If the user wants to get the prediction of a stock, the user again selects the stock from the list,clicks the “Prediction” button. There are two ways in which the system works: one is using theWeb Services and the other one is using the Indicators which can be invoked by the user.Based on the method selected (Web Services or the Indicators), data is again fetched from thedatabase and the desired results are sent back to the user through the services level.The database is kept up to date with the latest data and this data is inserted into the databaseevery 15 minutes using the internet resources. After every 15 minutes, new data is inserted intothe database and the old one is removed. Google stocks (http://finance.google.com) are being

16 P a g eused for fetching the data for our system and the programming code keeps on updating thedatabase periodically.2.3 Use Cases:Fig3: Use Case Diagram

17 P a g eUse Case Description:2.3.1 Use Case Casual Description

18 P a g e2.3.2 Functional Specification Requirement:

19 P a g e

20 P a g e

21 P a g e

22 P a g e2.3.3 Interaction Diagrams of key Use Cases:Fig 4: Use Case #2 and #3 Interaction Diagram

23 P a g eFig 5: Use Case #5 Interaction Diagram

24 P a g eFig 6: Use Case #7 Interaction DiagramNotes: The diagram starts from the controller. The controller gets a particular stock(s) andpasses it to the Predictor. The predictor may call the web services or use the indicators to getthe prediction based on the unique name and ID of the stock. Data is fetched from the storage,calculations are performed and finally again sent back to the storage and the final result: buyor sell are sent to the displayer which displays it on the user interface.

25 P a g e2.4 Activity Diagrams:Use Case #1 (Manage Account):Fig 7: Use Case #1 Activity Diagram (Manage Account)Steps:a. Administrator enters into the home page.b. Fills in the user name and password.

26 P a g ec. If the username/

Stock market prediction is the act of trying to determine the future value of a company stock or other financial instrument traded on a financial exchange. The successful prediction of a stock's future price could yield significant profit. The stock market is not an efficient market.

Related Documents:

Slicing Steaks 3563 Beef Tender, Select In Stock 3852 Angus XT Shoulder Clod, Choice In Stock 3853 Angus XT Chuck Roll, Choice 20/up In Stock 3856 Angus XT Peeled Knuckle In Stock 3857 Angus XT Inside Rounds In Stock 3858 Angus XT Flats, Choice In Stock 3859 Angus XT Eye Of Round, Choice In Stock 3507 Point Off Bnls Beef Brisket, Choice In Stock

monthly cycle for medium- to long-term issues. This dichotomy facilitates responsiveness in short-term planning while freeing senior management to focus on high-impact decisions over the longer-term horizon (and only by exception on operational issues). Sequential process steps. Both the weekly and monthly cycles follow a

36th International Symposium on Forecasting in Santander, Spain, a magnificent city in the northern region of Cantabria. This is the second time that the Symposium takes place in the Magdalena Palace, a majestic site hosting once more a truly international gathering of academic and industry forecasters.

Stock Market Game Time: 15 Minutes Requires: StockTrak Curriculum , Computer Access Buying Your First Stock This lesson is an introduction to buying a stock. Students will be introduced to basic vocabulary that is involved with a buying and owning a stock. Stu-dents will be going through the entire process of buying a stock from looking

Keywords: Forecastability, Stock returns, Non-linear models, Efficient markets. 1. Introduction: Random walk theory For reasons that are probably obvious, stock market prices have been the most analysed eco- nomic data during the past forty years or so. The basic question most asked is - are (real) price

Mastercam Training Guide Mill-Lesson-4-9 6. Change the parameters to match the Stock Setup screenshot below: Stock Setup Stock Origin The stock origin is the X-Y-Z coordinate position of the point indicated by the cross in the picture of the stock model. Use it so Mastercam knows where your stock model is located relative to your part and

WPX Energy, Inc. Common Stock We are offering 27,000,000 shares of our common stock. Our common stock is listed on the New York Stock Exchange under the symbol “WPX.” On July 10, 2015, the last reported sale price for our common stock on the New York Stock Exchange (the “NYSE”) was 11.22 per share.

The Alex Rider series is a fast-paced and action-packed set of books, ideal for lovers of spies and action. Teen readers will adore this unforgettable and enthralling series. Tomasz Hawryszczuk, age 9 This series of books is a must read for anyone over the age of 9 who likes spy stories, gadgets and danger. Best books ever! The Alex Rider series is an amazing set of books based on a 14 year .