Quantitative Investment Strategies with Python

Overview

Quantitative investment strategies are used by mutual funds, hedge funds and investors across all asset classes to identify the most attractive investment opportunities. The aim is to remove any qualitative or emotional component from the investment decision making process.

The strategies vary from the relatively simple such as picking the highest yielding stocks to the very complex which may involve optimising over many variables and markets. 

When well implemented these algorithms can generate consistent and attractive risk-adjusted returns.

Instructors

  • Dr. Brian Healy
  • Dr. Nick Firoozye

Venue

Level39 of One Canada Square, Canary Wharf, London’s premier fintech hub.

Registration

https://ai.thalesians.com/product/quantitative-investment-strategies-with-python-7-8-february-2019/

Schedule

TimeDay 1Day 2
08:30
09:00
Registration and welcome,
a tour of Level39
Registration and welcome
09:00
9:45
Lecture 1: What is a Quantitative
Strategy and who uses them?
Overview of the current state of the market.
Lecture 1: Market making strategies
9:45
10:30
Lecture 2: Overview of the
common quantitative strategies
Lecture 2: Carry Strategies
10:30
11:45
TutorialTutorial
11:45
12:00
Coffee BreakCoffee Break
12:00
12:45
Lecture 3: Data sources and preparationLecture 3: Fixed Income relative value
12:45
13:30
Lecture 4: Trend following
and momentum strategies
Lecture 4:Event driven strategies in equity markets
13:30
14:15
TutorialTutorial
14:15
15:00
LunchLunch
15:00
15:45
Lecture 5: Factor investing in equitiesLecture 5: Mixing strategies and diversification
15:45
16:30
Lecture 6: Mean reversion strategiesLecture 6: Performance measurement 
16:30
17:15
TutorialTutorial
17:15
18: 15
LabLab

Syllabus

Investment Strategies

  • What types of strategy do firms use
  • Passive vs active strategies
  • Short term vs long terms strategies
  • What is a quantitative strategy 

Data and Preparation

  • What types of data do we use
  • Where does the data come from
  • Intraday vs lower frequency date
  • What is involved in cleaning the data
  • Centering, standardising and normalising data
  • Some common data pitfalls
  • Diagnostics

Model and feature selection

  • Model selection
  • Cross-validation: leave-N-out, K-fold
  • Cross-validation for time series
  • Sliding window for time series
  • Bootstrap
  • Ridge regression, L2 regularisation
  • LASSO regression, L1 regularisation

Trend following and momentum strategies

  • Outlier detection
  • Parameter estimation
  • Prediction
  • How many data points do we need?
  • Optimisation 
  • Long-Short strategies 
  • CTAs

Factor investing in equities

  • Fundamental factors
  • Fama-French and friends
  • Linear regression and ordinary leasts squares 
  • Residual diagnostics
  • Adjusted coefficient of determination
  • The F-statistic
  • The p-value
  • PCA methods – uses and interpretation

Mean reversion strategies

  • Time series analysis 
  • Autoregressive and moving average processes
  • Stationarity
  • Parametric tests
  • In-sample diagnostics
  • Pairs trading
  • Time series cross-validation
  • Predicting events

Market making strategies

  • Order books
  • Limits and order types
  • Liquidity and market impact
  • Machine learning approaches 

Carry Strategies

  • What is carry?
  • How is it measured?
  • Curve strategies in fixed income
  • Carry in the FX market
  • Volatility carry

Fixed Income relative value

  • Building a curve
  • Rich versus cheap
  • Cheapest to deliver and special bonds
  • Curve options
  • Short gamma

Event driven strategies in equity markets

  • Dividend capture and corporate events
  • M&A driven trades
  • What happens between events
  • Sizing the trade

Mixing strategies and diversification

  • The covariance matrix
  • Key properties of covariance matrices
  • Sample covariance matrix
  • The correlation matrix
  • The sample correlation matrix
  • Multicollinearity

Performance measurement

  • Performance terminology
  • P-hacking
  • The optimisation problem
  • Information ratios
  • Overfitting
  • Underfitting

Bibliography

Your course is designed to be self-contained. However, if you would like to read up on the content before, during, and/or after the course, we recommend the following books: