Scribd Logo
Upload

Welcome to Scribd!

  • 38 views

    Stock Price Trend Forecasting Using Supervised Learning Methods

    Uploaded by

    SHAIK AZAJUDDIN 18MIS7192
    The document summarizes research on using supervised learning methods to predict stock prices based on past returns and news indicators. It describes preprocessing stock market data from 500 S&P 500 companies, feature selection to identify important variables, adding a Twitter sentiment score feature, and applying regression models like gradient boosting, bagging, random forest and K-neighbors. Bagging and gradient boosting performed best. The goal is to construct a diversified stock portfolio to manage risk and maximize profit through predictive analysis of financial market data.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Stock Price Trend Forecasting Using Supervised Learning Methods

    Uploaded by

    SHAIK AZAJUDDIN 18MIS7192
    38 views17 pages
    The document summarizes research on using supervised learning methods to predict stock prices based on past returns and news indicators. It describes preprocessing stock market data from 500 S&P 500 companies, feature selection to identify important variables, adding a Twitter sentiment score feature, and applying regression models like gradient boosting, bagging, random forest and K-neighbors. Bagging and gradient boosting performed best. The goal is to construct a diversified stock portfolio to manage risk and maximize profit through predictive analysis of financial market data.

    Original Description:

    This paper is based on Stock Price Prediction

    Original Title

    StockPricePrediction

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document summarizes research on using supervised learning methods to predict stock prices based on past returns and news indicators. It describes preprocessing stock market data from 500 S&P 500 companies, feature selection to identify important variables, adding a Twitter sentiment score feature, and applying regression models like gradient boosting, bagging, random forest and K-neighbors. Bagging and gradient boosting performed best. The goal is to construct a diversified stock portfolio to manage risk and maximize profit through predictive analysis of financial market data.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    38 views17 pages

    Stock Price Trend Forecasting Using Supervised Learning Methods

    Uploaded by

    SHAIK AZAJUDDIN 18MIS7192
    The document summarizes research on using supervised learning methods to predict stock prices based on past returns and news indicators. It describes preprocessing stock market data from 500 S&P 500 companies, feature selection to identify important variables, adding a Twitter sentiment score feature, and applying regression models like gradient boosting, bagging, random forest and K-neighbors. Bagging and gradient boosting performed best. The goal is to construct a diversified stock portfolio to manage risk and maximize profit through predictive analysis of financial market data.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 17

    Stock Price Trend Forecasting using

    Supervised Learning Methods

    Saurabh Jain (201301128)


    Sharvil Katariya (201301129)
    Problem Statement
    The aim of the project is to examine a
    number of different forecasting techniques
    to predict future stock returns based on
    past returns and numerical news indicators
    to construct a portfolio of multiple stocks in
    order to diversify the risk. We do this by
    applying supervised learning methods for
    stock price forecasting by interpreting the
    seemingly chaotic market data.
    Motivation

    The fluctuation of stock market is violent and there are many complicated
    financial indicators. However, the advancement in technology, provides an
    opportunity to gain steady fortune from stock market and also can help experts
    to find out the most informative indicators to make better prediction. The
    prediction of the market value is of paramount importance to help in maximizing
    the profit of stock option purchase while keeping the risk low.
    1. Preprocessing and Cleaning

    2. Feature Extraction

    3. Twitter Sentiment Analysis and Score

    The Solution 4. Data Normalization

    5. Analysis of various supervised learning


    methods

    6. Conclusions
    Structure of Data

    We have taken these Historical Daily Stock Market Data for 500 companies of
    S&P 500 Market since the company came into existence.

    The base features in this data that we collected were :

    Open, High, Low, Close, Volume, Adjusted Close for each company and S&P
    index.
    Feature Selection
    To prune out less useful features, in Feature Selection, we selected features according to
    the k highest scores, with the help of a linear model for testing the effect of a single
    regressor, sequentially for many regressors. (SelectKBest Algorithm, with f_regression as
    the scorer used)

    This is done in 3 steps:


    1. Start with a constant model, M0
    2. Try all models M1 consisting of just one feature and pick the best according to the F
    statistic.
    3. Try all models M2 consisting of M1 plus one other feature and pick the best …

    A F-test (Wikipedia) is a way of comparing the significance of the improvement of a model, with
    respect to the addition of new variables.
    Twitter Sentiment Score as new feature

    Social media plays important role in predicting the stock market return values.

    So, we then appended our data with one more feature -

    Twitter’s Daily Sentiment Score for each company based upon the user’s tweets
    about that particular company and also the tweets on that company’s page.

    Once we were ready with complete set of features, we normalized our data for
    better results.
    Models Used

    We have used the following regression models for forecasting and comparison :

    1. Neural Networks
    2. Bagging Regressor
    3. Random Forest Regressor
    4. Gradient Boosting Regressor
    5. Adaboost Regressor
    6. K Neighbour Regressor
    Results : AdaBoost Regressor

    American Airlines Group (AAL)

    R-squared : 2.988 e-07

    RMSE : 0.909
    Results : Gradient Boosting Regressor

    American Airlines Group (AAL)

    R-squared : 1.275 e-07

    RMSE : 0.961
    Results : Bagging Regressor
    American Airlines Group (AAL)

    R-squared : 1.329 e-07

    RMSE : 0.959
    Results : K Neighbours Regressor
    American Airlines Group (AAL)

    R-squared : 0.00039

    RMSE : Too High


    Results : Random Forest Regressor

    American Airlines Group (AAL)

    R-squared : 1.432 e-07

    RMSE : 0.956
    Results

    ● Gradient Descent Regressor: Gradient Descent + Boosting


    ● Bagging Regressor is found to perform well as Bagging (Bootstrap sampling)
    relies on the fact that combination of many independent base learners will
    significantly decrease the error.
    ● Therefore we want to produce as many independent base learners as
    possible.
    ● Each base learner is generated by sampling the original data set with
    replacement.
    Bagging vs Boosting
    Bagging:

    1. Parallel ensemble: Each model is built independently


    2. Aim to decrease variance, not bias
    3. Suitable for high variance low bias models (complex models)
    4. An example of a tree based method is random forest, which develop fully grown trees (note that
    RF modifies the grown procedure to reduce the correlation between trees)

    Boosting:

    1. Sequential ensemble: try to add new models that do well where previous models lack
    2. Aim to decrease bias, and not variance
    3. Suitable for low variance high bias models
    4. An example of a tree based method is gradient boosting
    Milestones

    18 March 2016 12 April 2016


    Collection and Preprocessing of Implementation of various
    data, Feature Extraction and regression models and
    Data Normalization classification based upon the
    result.

    30 March 2016 15 April 2016


    Include Social Media Sentiment Comparison and Analysis
    Analysis as a feature to predict stock of Different Models
    trend more accurately.
    Thank You

    You might also like