Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Application of Machine Learning in Stock Market Forecasting: A Case Study of Disney Stock

Dengxin Huang

TL;DR

This work addresses predicting Disney stock returns by combining classic asset-pricing factors with data-driven feature engineering in a 750×16 dataset. It systematically compares Fama French 3-factor modeling, linear regression, random forest, and gradient boosting, after rigorous data cleaning, preprocessing, and clustering-based feature construction. The key finding is that linear regression using common factors and a clustering-derived idiosyncratic feature achieves the highest accuracy ($95.23\%$), outperforming nonparametric ensemble methods. The study demonstrates the practical value of integrating established financial factors with machine learning features for stock return prediction and suggests expanding the analysis to more stocks and longer periods to enhance generalizability.

Abstract

This document presents a stock market analysis conducted on a dataset consisting of 750 instances and 16 attributes donated in 2014-10-23. The analysis includes an exploratory data analysis (EDA) section, feature engineering, data preparation, model selection, and insights from the analysis. The Fama French 3-factor model is also utilized in the analysis. The results of the analysis are presented, with linear regression being the best-performing model.

Application of Machine Learning in Stock Market Forecasting: A Case Study of Disney Stock

TL;DR

This work addresses predicting Disney stock returns by combining classic asset-pricing factors with data-driven feature engineering in a 750×16 dataset. It systematically compares Fama French 3-factor modeling, linear regression, random forest, and gradient boosting, after rigorous data cleaning, preprocessing, and clustering-based feature construction. The key finding is that linear regression using common factors and a clustering-derived idiosyncratic feature achieves the highest accuracy (), outperforming nonparametric ensemble methods. The study demonstrates the practical value of integrating established financial factors with machine learning features for stock return prediction and suggests expanding the analysis to more stocks and longer periods to enhance generalizability.

Abstract

This document presents a stock market analysis conducted on a dataset consisting of 750 instances and 16 attributes donated in 2014-10-23. The analysis includes an exploratory data analysis (EDA) section, feature engineering, data preparation, model selection, and insights from the analysis. The Fama French 3-factor model is also utilized in the analysis. The results of the analysis are presented, with linear regression being the best-performing model.
Paper Structure (18 sections, 4 equations, 7 figures)

This paper contains 18 sections, 4 equations, 7 figures.

Table of Contents

  1. Introduction
  2. Exploratory Data Analysis
  3. Data Cleaning, Preprocessing, Feature Engineering, and Clustering
  4. Data Cleaning
  5. Preprocessing
  6. Feature Engineering
  7. Clustering
  8. Model Selection
  9. Fama French 3-factor model
  10. Linear Regression
  11. Random Forest Regressor
  12. Gradient Boosting Regression
  13. Modeling
  14. Linear Regression
  15. Random Forest
  16. ...and 3 more sections

Figures (7)

  • Figure 1: Time series plot showing the volatility of the stock prices over time.
  • Figure 2: Frequency histograms for various columns.
  • Figure 3: Stock prices after log transformation.
  • Figure 4: Clustering analysis used to identify stocks that fluctuate with Disney stock.
  • Figure 5: Accuracy of Linear Regression Model
  • ...and 2 more figures