The Wall Street Quants BootCamp

We guide you through a 30,000 foot tour of quant research & trading. Develop a high-level understanding of key concepts before we dive into the fine details.

Sub-topics: backtesting, overfitting, idea generation, types of quant strategies, the major quant firms

Learn the most widely used coding language by quants today: Python. We teach it from the ground up with attention to the specific skills & libraries desired by quant firms. There are lots of practical code examples (focusing on quant applications) and interactive exercises designed to bring you up to speed even if you don’t have a coding background.

Sub-topics: python, numpy, pandas, data analysis, data sources, plotting

Quants have a rigorous framework for deciding if a strategy is any good. You can’t manage what you don’t measure! Learn the current industry-standard techniques & metrics used in strategy evaluation.

Sub-topics: sharpe ratios, regressions, volatility, returns, drawdowns, alpha, beta

Mature quant models are often combinations of many (sometimes hundreds!) underlying strategies. Understand best-practices for building powerful quant models by optimally combining individual strategies.

Sub-topics: correlation, breadth, covariance matrix, volatility-weighting, risk-parity, sharpe ratio – weighting, generalized weighting, robust weighting

You’ve found a dataset / signal that you think predicts returns. Now, how do you robustly and accurately express this signal as a tradable portfolio of securities?

Sub-topics: signal transformations, cross-sectional (XS) models, time-series (TS) models, implementation

Backtesting, the art & science of simulating your strategy with historical data, is the bread and butter of quant. We teach the two types of backtesting used in the quant workflow.

Sub-topics: unconstrained backtests, optimized backtests

Conducting “alpha” research to discover new strategies is an art! Learn industry best-practices and how to avoid common pitfalls (including overfitting).

Sub-topics: overfitting, out of sample testing, priors, model complexity, the utility of machine learning

Learn the main strategy types used in industry today and leave with a powerful arsenal! There’s a focus on stronger “statistical arbitrage” strategies. We also teach methods/frameworks for improving strategies and taking them to the next level.

Sub-topics: event-driven, momentum, reversal, pairs trading, statistical arbitrage, value, crypto strategies

We guide you through a project that mimics the actual work you’d do at a quant hedge fund.

The goal is to develop a statistical arbitrage strategy in cryptocurrencies. We provide a research outline for the project, including a framework for building profitable mean-reversion & momentum strategies. Ambitious students can deploy their strategy on real money using code provided at the end of the course in Part 13.

Trading costs can make or break a strategy, especially for higher frequency “statistical arbitrage” strategies. Learn the main types of trading costs and how to incorporate them in backtests. We further explore transaction cost modeling and methods for minimizing costs.

Sub-topics: commissions, slippage, bid-ask spread, turnover, transaction cost modeling, portfolio optimization

Understand different risk sources (“factors”) beyond the market (“beta”), how to measure them and how to immunize (“neutralize”) your portfolio to them to build a more robust strategy.

Sub-topics: risk factors, barra risk models, fama-french factor models, beta/loading, risk contribution, risk-neutralization

Most quants use an optimization as the “final step” in forming portfolios. Optimizers strategically balance return, risk, transaction costs and portfolio constraints (IE. long-only). Learn how to implement one from scratch using convex-optimization and the cvxpy library.

Sub-topics: mean-variance optimization, convex optimization, cvxpy, transaction cost models, portfolio constraints

We wrap up the course by walking you through the code for a flexible python trading library that allows for live trading (and backtesting) fully-automated quant strategies on real money. Learn how to implement a robust end-to-end production trading system, and impress employers by building a live personal track-record.

Sub-topics: order management, live data feeds, broker APIs, logging