Syllabus for Fall 2021 Course home
- This is definitely a well done class; take this class before you get into other AI/ML classes. You will learn a lot from here that will make your life easy for other classes.
- You will learn numpy, pandas, data cleaning and visualization in this course
Lecture is freely available for anyone!
https://omscs.gatech.edu/cs-7646-machine-learning-trading-course-videos
Hello Numpy and pandas!
- Reading and plotting stock data: 1-1.ipynb
- Working with multiple stocks: 1-2.ipynb
- Power of NumPy: 1-3.ipynb
- Statistical analysis of time series: 1-4.ipynb
Optimizations
- Incomplete data: 1-5.ipynb
- Histograms and scatter plots: 1-6.ipynb
- Histograms and scatter plots
- A closer lok at daily returns
- What would it look like?
- Histogram of daily returns
- How to plot a histogram
- Computing histogram statistics
- Compare two histograms
- Plot two histograms together
- Scatter plots
- Fitting a line to data points
- Slope != correlation
- Correlation vs slope
- Scatter plots in python
- Real world use of kurtosis
- Sharpe ratio and other portfolio statistics: 1-7.ipynb
- Overview
- Daily Portfolio values
- Portfolio Statistics
- Which portfolio is better?
- Sharpe Ratio
- Form of the Sharpe Ratio
- Computing Sharpie Ratio
- Optimizers: Building a parameterized model: 1-8.ipynb
- What is an optimizer?
- Minimization example
- Minimizer in pythong
- How to defeat a minimizer
- Convex Problems
- Building a parameterized model
- What is a good error metric?
- Minimizers finds coefficients
- Fit a line to given data points
- And it works for polynomials too!
Intro to ML
-
Optimizers: How to optimize a portfolio: 1-9.ipynb
- What is portfolio optimization
- The difference optimization can make
- Which criteria is easiest to solve for?
- Framing te problem
- Ranges and constraints
-
How Machine Learning is used at a hedge fund: 3-1.ipynb
- How Machine learning is used at a hedge fund overview
- The ML problem
- What's X and Y
- Supervised Regression Learning
- How it works with stock data
- Example at a fintech company
- Price forecasting demo
- Backtesting
- ML tool in use
- Problems with regression
-
- Regression Introduction
- Parametric Regression
- K Nearest Neighbor
- How to predict
- Kernel Regression
- Parametric vs Non-parametric
- Training and testing
- Example for linear Regression
-
Assessing learning Algorithm: 3-3.ipynb
- Fitting data
- RMS Error
- In sample vs out of sample
- Cross validation, Roll forward cross validation
- Correlation and RMS Error
- Overfitting
-
Ensemble learners - bagging and boosting: 3-4.ipynb
- Ensemble
- Bootstrap aggregating - bagging
- Overfitting
- Bagging example
- Boosting
- Overfitation
-
Decision Tree & Random forest: 3-5_decision_tree.ipynb
- 2 lecture videos
- 1st part: what does the tree look like
- 2nd part: building the tree
- JR Quinlan's algorithm
- A Cutler Algorithm
- Random forest
- 2 lecture videos
-
So you want to be a hedge fund manager? 2-1.ipynb
- Types of funds
- Liquidity and capitalization
- What type of fund is it?
- Incentives for fund managers
- Two and twenty
- How funds attract investors
- Hedge fund goals and metrics
- The computing inside a hedge fund
-
- Market Mechanics Overview
- What is in an order?
- The order book
- Up or down?
- How orders affect the order book?
- How orders get to the exchanges
- How hedge funds exploit market mechanics
- additional order types
- Mechanics of short selling: Entry
- Short Selling
- Mechanics of short selling: Exit
- What can go wrong?
-
What is a company worth? 2-3.ipynb
- What is a company worth?
- Why company value matters?
- The Balch Bond
- The value of a future dollar
- Intrinsic Value
- Book Value
- Market Capitalization
- Why information affects stock price
- Compute Company value
- Would buy this stock?
-
The Capital Assets # Model (CAPM): 2-4.ipynb
- The Capital Assets # Model
- Definition of a portfolio
- Portfolio return
- The market portfolio
- The CAPM Equation
- Compare alpha and beta
- CAPM vs active management
- CAPM for portfolios
- Implications of CAPM
- Arbitrage # Theory
-
How hedge funds use the CAPM: 2-5.ipynb
- Risks for hedge funds
- Two stock scenario
- Two stock CAPM Math
- Allocations remove market risk
- How does it work
- CAPM for hedge funds Summary
-
- Technical versus fundamental analysis
- Characteristics
- Potential indicators
- When is technical analysis valuable?
- A few indicators:
- Momentum
- Simple Moving Average
- Bollinger bands
- Buy or Sell?
- Normalization
-
- Overview
- How data is aggregated
- Price anomaly
- Stock Splits
- Split adjustment
- Dividends
- Adjusting for Dividends
- Survivor bias
-
Efficient Markets Hypothesis: 2-8.ipynb
- Our hypothesis
- EMH hypothesis
- Origin of Information
- 3 forms of EMH
- The EMH Prohibits
- Is the EMH correct?
- Time Series Data
- Vectorize me PPT
- Both are covered in the youtube video