📊 Stock Market Classification (ISLP Chapter 4 Labs)

This repository contains my practice notebooks from An Introduction to Statistical Learning with Python (ISLP), Chapter 4 labs.
The focus is on classification methods for predicting stock market movements.

---

📌 Topics Covered

1. Logistic Regression

   • Predicting Direction (Up/Down) using lagged returns
   • Interpreting coefficients
   • Probability outputs vs class predictions

2. Linear Discriminant Analysis (LDA)

   • Assumes groups are normally distributed with equal variance
   • Finds linear boundaries between classes

3. Quadratic Discriminant Analysis (QDA)

   • More flexible than LDA
   • Allows curved decision boundaries

4. Naive Bayes

   • Simple probabilistic classifier
   • Assumes predictor independence
   • Often works surprisingly well

---

📊 Dataset

We use the Smarket dataset, included in the ISLP library:

• Daily % returns for the S&P 500 (2001–2005)
• Predictors: Lag1, Lag2, ..., Lag5, Volume
• Response: Direction (Up/Down)

---

🚀 Methods

• Split data into training (2001–2004) and test (2005) sets
• Fit models on training set
• Evaluate on test set using confusion matrices and accuracy

---

🛠 Tools & Libraries

• Python (3.9+)
• pandas, numpy
• statsmodels
• scikit-learn
• matplotlib
• ISLP package

---

💻 How to Run

1. Clone this repo:

   git clone https://github.com/Oxcheatcode/logistic-regression.git
   cd islp-logistic-classification

2. pip install -r requirements.txt 3.jupyter notebook

🌟 Key Learnings

Logistic regression outputs probabilities (great for classification)

LDA assumes linear separation; QDA allows non-linear boundaries

Naive Bayes is simple yet effective

Always evaluate with training vs test data to avoid overfitting

🔗 References

An Introduction to Statistical Learning with Python

ISLP package