■ Month 1 – Foundations + ML Kickstart
Goal: Learn Python, basic stats, and start using ML for simple trading signals.
■ Math & Stats
• Mean, variance, std, skew, kurtosis.
• Probability basics: random variables, expected value.
• Normal vs heavy-tailed distributions (important for returns).
• Hands-on: Load stock data (yfinance), compute returns, volatility, Sharpe ratio, plot histogram of
returns vs normal distribution.
■ Python & Tools
• Pandas, NumPy, Matplotlib basics.
• Scikit-learn intro: how to use fit() and predict().
■ Machine Learning (Early Start)
• Regression (linear & logistic) → predicting returns/directions.
• Train/test split & cross-validation.
• Feature engineering: Moving averages, Rolling volatility, Momentum (past returns).
• Hands-on: Build dataset, train logistic regression, backtest toy strategy.
■ Finance Basics
• What is a return (simple vs log)?
• Risk/Reward: Sharpe, Sortino.
• Intro to trading costs (slippage, commissions).
■■ Practice Project (End of Month 1)
• “My First ML Trading Model”.
• Pull 1 year of SPY data.
• Engineer features: 5-day MA, 10-day MA, 20-day volatility, 5-day momentum.
• Train/test split → logistic regression to classify next-day return sign.
• Build simple strategy: Long if model says “up,” flat if “down.”
• Evaluate: accuracy, Sharpe ratio, max drawdown.
■ Month 2 – Stats Expansion + ML Deeper
Goal: Strengthen probability/stat inference and build more robust ML models.
■ Math & Stats
• Random variables, expected value, Central Limit Theorem (CLT).
• Hypothesis testing, confidence intervals, p-values.
• Multiple regression analysis.
• Hands-on: Test if mean returns are significantly > 0.
■ Python & Tools
• Scikit-learn deeper: preprocessing, feature scaling.
• Cross-validation methods.
■ Machine Learning
• Multiple regression & logistic regression with multiple features.
• Hands-on: Predict stock direction using rolling averages, volatility.
■ Finance Basics
• Sharpe vs Sortino ratio in practice.
• Risk-adjusted return comparison.
■■ Practice Project (End of Month 2)
• Predict stock direction using logistic regression with multiple features.
• Backtest with train/test split and rolling validation.
• Compare risk-adjusted performance to baseline momentum strategy.
■ Month 3 – Correlation & Dependence
Goal: Learn how assets move together and test mean-reversion vs momentum.
■ Math & Stats
• Covariance, correlation, correlation matrix.
• Autocorrelation & stationarity in time series.
• Hands-on: Build correlation heatmap for assets.
■ Python & Tools
• Statsmodels: autocorrelation, ADF tests.
■ Machine Learning
• Decision trees vs regression models.
• Feature importance analysis.
■ Finance Basics
• Portfolio theory basics (Markowitz efficient frontier).
• Diversification benefits.
■■ Practice Project (End of Month 3)
• Backtest mean-reversion vs momentum strategies.
• Compare Sharpe, Sortino, max drawdown.
• Visualize strategy performance side by side.
