Gen AI and Quantitative Investments
led by Kerry Back
Week 1
August 25–27
TUE AUG 25
Getting Started
Course mechanics. Intro to quant investing. Intro to gen AI. Our AI Lab (Claude Code/GLM 4.7 + Python in a container).
THU AUG 27
Median of Peer Multiples
Introduction to the Nasdaq Data Link data. Finding industry/size peers and valuing by multiples. Analyzing model/market differences.
·Files
Week 2
September 1–3
TUE SEP 1
Trading on Predictions
Turning a valuation signal into a portfolio. Ranking stocks, going long and short, and backtesting what the strategy would have earned.
Slides·Files
THU SEP 3
Linear Regression for Multiples
Predicting a multiple from a company's fundamentals. Reading the regression output, holding out a test sample, and the transformations that decide the answer.
·Files
Week 3
September 8–10
TUE SEP 8
Ridge, Lasso, and Trees
Penalized regression, regression trees, and gradient boosting, each with a slider on the slide. Overfitting, and choosing a tuning parameter by cross-validation.
Week 4
September 15–17
TUE SEP 15
Build a Stock Recommender App
Wrapping the fitted models in a web app. The models predict returns, and the app turns those return predictions into buy and sell recommendations.
THU SEP 17
Build a DCF App
Build a web app that retrieves historical financials and generates a two-stage DCF enterprise valuation, allowing the user to input key values or to extrapolate historical trends.
Week 5
September 22–24
TUE SEP 22
Factor ETFs
Explore the construction and performance of the iShares/MSCI factor ETFs.
THU SEP 24
Analyzing Earnings Calls
Where transcripts come from and what they cost. Can a model read a call and tell you to buy, and how fast would you have to act? Four architectures, and what we already know.
Week 6
September 29 – October 1
TUE SEP 29
Analyzing News Headlines
A corpus of 1.3 million headlines, and what is left of it once you check the coverage. Scoring a headline blind to the price, then asking whether the score knew anything.