Stock Prediction with CAPM and Fama-French Model (#418)

## Pull Request for PyVerse 💡 ### Requesting to submit a pull request to the PyVerse repository. --- #### Issue Title Stock Prediction with CAPM and Fama-French Model - [x] I have provided the issue title. --- #### Info about the Related Issue **What's the goal of the project?** The goal of this project is to predict the stock returns of Mahindra using the Capital Asset Pricing Model (CAPM) and the Fama-French Three-Factor Model. The project aims to compare the accuracy of both models and determine which is better at predicting the company's stock returns based on historical financial data. - [x] I have described the aim of the project. --- #### Name Sharayu Anuse - [x] I have provided my name. --- #### GitHub ID 114616759 - [x] I have provided my GitHub ID. --- #### Email ID sharayu.anuse@gmail.com - [x] I have provided my email ID. --- #### Identify Yourself **Mention in which program you are contributing (e.g., WoB, GSSOC, SSOC, SWOC).** GSSOC-Ext, Hacktoberfest - [x] I have mentioned my participant role. --- #### Closes Closes: #250 - [x] I have provided the issue number. --- #### Describe the Add-ons or Changes You've Made **Give a clear description of what you have added or modified.** - Implemented both the CAPM and Fama-French Three-Factor Model for predicting stock returns. - Added functionality for data analysis using Pandas and NumPy. - Plotted visualizations to compare the predicted returns with actual returns using Matplotlib. - Provided statistical analysis to demonstrate that the Fama-French model shows 36% greater accuracy than CAPM. - [x] I have described my changes. --- #### Type of Change **Select the type of change:** - [ ] Bug fix (non-breaking change which fixes an issue) - [x] New feature (non-breaking change which adds functionality) - [ ] Code style update (formatting, local variables) - [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected) - [ ] This change requires a documentation update --- #### How Has This Been Tested? **Describe how your changes have been tested.** - Generated accuracy scores to compare the two models. - Visualization Testing: Plotted data to visually verify the accuracy of predicted vs. actual stock returns. - [x] I have described my testing process. --- #### Checklist **Please confirm the following:** - [x] My code follows the guidelines of this project. - [x] I have performed a self-review of my own code. - [x] I have commented my code, particularly wherever it was hard to understand. - [x] I have made corresponding changes to the documentation. - [x] My changes generate no new warnings. - [x] I have added things that prove my fix is effective or that my feature works. - [x] Any dependent changes have been merged and published in downstream modules.
UTSAVS26 · Oct 16, 2024 · 9b156a1 · 9b156a1
2 parents b5596c4 + 000d497
commit 9b156a1
Show file tree

Hide file tree

Showing 14 changed files with 1,806 additions and 0 deletions.
diff --git a/Data_Science/stock_prediction/CAPM_FamaFrench.ipynb b/Data_Science/stock_prediction/CAPM_FamaFrench.ipynb
diff --git a/Data_Science/stock_prediction/README.md b/Data_Science/stock_prediction/README.md
@@ -0,0 +1,79 @@
+## Stock Prediction Using CAPM and Fama-French Three-Factor Model
+
+### 🎯 **Goal**
+
+The main goal of this project is to predict the stock returns for Mahindra Company using two financial models: the Capital Asset Pricing Model (CAPM) and the Fama-French Three-Factor Model. The project aims to compare the accuracy and performance of both models in determining the expected stock returns based on historical data.
+
+### 🧵 **Dataset**
+
+The historical stock data for Mahindra Company has been sourced from Yahoo Finance.
+
+### 🧾 **Description**
+
+This project leverages two well-known models in finance, CAPM and the Fama-French Three-Factor Model, to predict stock returns. The project uses Mahindra’s historical stock data to evaluate how each model performs in explaining the stock's return based on market factors.
+
+The Capital Asset Pricing Model (CAPM) considers only the market risk (Beta) to determine returns, while the Fama-French model incorporates three factors: market risk, size of the firm (SMB), and value vs. growth (HML).
+
+### 🧮 **What I had done!**
+
+- Data Collection: Retrieved Mahindra’s historical stock data from Yahoo Finance.
+- Data Preprocessing: Cleaned the data and calculated necessary variables like excess returns, risk-free rate, and market factors.
+- CAPM Implementation:
+   - Estimated the beta of Mahindra’s stock.
+   - Applied the CAPM formula to predict expected returns.
+- Fama-French Three-Factor Model:
+   - Retrieved and incorporated additional factors: SMB (Small Minus Big) and HML (High Minus Low).
+   - Predicted expected returns using the three-factor model.
+- Model Evaluation:
+   - Calculated statistical measures like R-squared and p-values to evaluate the models' accuracy.
+   - Compared the results of the CAPM and Fama-French models.
+- Visualization: Generated visual plots to represent the relationship between predicted and actual returns.
+
+### 🚀 **Models Implemented**
+
+- Capital Asset Pricing Model (CAPM): Chosen for its simplicity in predicting stock returns based on the market risk factor (Beta).
+- Fama-French Three-Factor Model: Selected because it extends CAPM by incorporating two additional factors, SMB (size factor) and HML (value factor), making it more comprehensive for predicting stock returns.
+
+- Why these models?
+
+CAPM is a fundamental model that provides a baseline understanding of stock returns based on market volatility.
+The Fama-French model was chosen for its greater accuracy in capturing returns by considering multiple factors.
+
+### 📚 **Libraries Needed**
+
+- Pandas: For data manipulation and analysis.
+- NumPy: To perform numerical operations on arrays.
+- Matplotlib: For visualizing data and results.
+- Statsmodels: For performing regression analysis and statistical computations.
+
+### 📊 **Exploratory Data Analysis Results**
+
+![output_12](https://github.com/user-attachments/assets/9966a03d-5a8c-4629-858f-eac88d7c51db)
+![output_11](https://github.com/user-attachments/assets/32097575-bf49-4c43-b46a-bfca9242fe19)
+![output_10](https://github.com/user-attachments/assets/2a181df5-fbf9-4e89-9077-3e4e9e694358)
+![output_9](https://github.com/user-attachments/assets/9b74abb2-9a7f-4148-805d-020f14618f45)
+![output_8](https://github.com/user-attachments/assets/9903c057-6d84-42a2-82da-320042380ad0)
+![output_7](https://github.com/user-attachments/assets/974b08eb-a69c-4d66-9fca-b124bff53203)
+![output_6](https://github.com/user-attachments/assets/720d30ed-53b6-4a56-9cd9-6330f920b472)
+![output_5](https://github.com/user-attachments/assets/def28983-ca3b-4de1-84e9-612c24cec65e)
+![output_4](https://github.com/user-attachments/assets/bb195ec1-ae23-4bec-9da8-275169dcd41d)
+![output_3](https://github.com/user-attachments/assets/6dae166b-725e-41d4-9d74-ee51f56b1384)
+![output_2](https://github.com/user-attachments/assets/f0622a64-6b1a-4f73-8a82-c05579623689)
+![output_1](https://github.com/user-attachments/assets/befa3864-12c0-4d69-961e-ab11f2e572b1)
+
+
+### 📈 **Performance of the Models based on the Accuracy Scores**
+
+- CAPM Accuracy: The model provided a decent explanation of stock returns but lacked accuracy in certain periods.
+- Fama-French Model Accuracy: Outperformed CAPM with an R-squared value that was 36% higher, showing greater accuracy in predicting returns.
+
+The Fama-French model clearly provided a more accurate prediction due to the inclusion of additional factors beyond just market risk.
+
+
+### 📢 **Conclusion**
+
+In conclusion, while the CAPM model offers a fundamental approach to predicting stock returns, the Fama-French Three-Factor Model is more effective for Mahindra's stock due to its additional factors, leading to a 36% improvement in accuracy. This project highlights the importance of using more comprehensive models like Fama-French for stock prediction tasks.odels for the particular projects.
+
+### ✒️ **Your Signature**
+
+Sharayu Anuse
diff --git a/Data_Science/stock_prediction/output_1.png b/Data_Science/stock_prediction/output_1.png
diff --git a/Data_Science/stock_prediction/output_10.png b/Data_Science/stock_prediction/output_10.png
diff --git a/Data_Science/stock_prediction/output_11.png b/Data_Science/stock_prediction/output_11.png
diff --git a/Data_Science/stock_prediction/output_12.png b/Data_Science/stock_prediction/output_12.png
diff --git a/Data_Science/stock_prediction/output_2.png b/Data_Science/stock_prediction/output_2.png
diff --git a/Data_Science/stock_prediction/output_3.png b/Data_Science/stock_prediction/output_3.png
diff --git a/Data_Science/stock_prediction/output_4.png b/Data_Science/stock_prediction/output_4.png
diff --git a/Data_Science/stock_prediction/output_5.png b/Data_Science/stock_prediction/output_5.png
diff --git a/Data_Science/stock_prediction/output_6.png b/Data_Science/stock_prediction/output_6.png
diff --git a/Data_Science/stock_prediction/output_7.png b/Data_Science/stock_prediction/output_7.png
diff --git a/Data_Science/stock_prediction/output_8.png b/Data_Science/stock_prediction/output_8.png
diff --git a/Data_Science/stock_prediction/output_9.png b/Data_Science/stock_prediction/output_9.png