Machine Learning 4 Trading [Talk]

Quantitative finance and programming trading strategies w/ Yves Hilpisch, The Python Quants


  • In case you’re new to this, Yves explains what it means to be a “quant” and how one tries to make sense of markets which varies from other types of participants.
  • How a quant looks for opportunities and areas where they can exploit the market. Plus, a brief walk through history; applying quantitative methods to finance.
  • Yves shares his thoughts on the effectiveness of machine learning (and similar techniques) and where he sees potential value from social sentiment data.
  • Many years ago, Yves started The Python Quants—he speaks about his motives and what goes on inside the events and meet-ups they host in various countries.
  • Out of all the programming languages in existence, Yves talks about why Python is his preferred choice for financial applications and it’s overall popularity.
  • Yves uses Pandas—a Python library created by a programmer within a large hedge fund—as a paramount example for open-sourced code and value of collaboration.
  • A practical way for non-programmers to learn how to code in Python, educational resources, and what sort of commitment is required to become proficient.

Strategy development—powered by machine learning


  • How large funds and institutions put on $100-million positions; how they work orders into the market, structure the trade and handle market impact etc.
  • Morgan explains why he feels as though the common approach to strategy development is counter intuitive, and shares an alternative 3-step formula.
  • A simple description of how machine learning and data science is being used by traders, and an example of how ML has been used to improve existing strategies.

Ứng dụng tốt (và không thật tốt) của Học máy trong kinh doanh và tài chính.


Học máy (Machine learning) là một chủ đề nóng hiện nay,  rất nhiều người tự hỏi làm thế nào nó có thể được sử dụng trong tài chính và kinh doanh. Nếu sử dụng ML một cách “ngây thơ”, nó có thể dẫn tới rất nhiều rủi ro. Chúng tôi sẽ thảo luận về lý do tại sao rủi ro có thể xảy ra và một số cách tốt để sử dụng ML một cách cẩn thận.

Chủ đề của cuộc thảo luận:

  • Máy học (Machine learning) là gì và làm thế nào nó được sử dụng trong cuộc sống hàng ngày?
  • Học có giám sát vs Học không có giám sát, và khi nào thì sử dụng chúng.
  • Liệu Học máy có cung cấp bất cứ điều gì nhiều hơn các phương pháp thống kê truyền thống.
  • Ứng dụng tốt (và không thật tốt) của máy học trong kinh doanh và tài chính.
  • Sự cân bằng giữa đơn giản và phức tạp.

Machine learning for algorithmic trading


  • Bert’s takeaway from reading a ton of trading books, and why there’s “no such thing as a bad book” – because you can always learn something (even if it’s doing the opposite).
  • In it’s purest form, Bert explains the purpose of machine learning, and gives an example of how it’s used in everyday technology.
  • How Bert uses machine learning to discover and create effective trading algorithms, from starting point through to live trading. And the information which is “nice to incorporate” outside of price.
  • The ways in which machine learning techniques differ from more common ways of developing algorithms, and how it removes further bias from your models.
  • Bert shares his opinion about whether a strategy should make sense logically, or if a statistical edge is the only evidence you need.
  • How Bert  thinks about diversification, and why he prefers to allocate additional capital to new markets, instead of adding to the markets he’s already actively trading.
  • The affects of machines becoming more and more prominent in the trading landscape; positive or negative?
  • Bert speaks about his attitude to always be learning and striving for continual progression. As well as his mindset of setting goals that are just out of reach, to really push yourself.
  • And much, much more…

Machine Learning With Kris Longmore


Machine learning has seen a huge amount of growth over recent years with the increase in available data and processing power.

It’s an incredibly powerful toolset for uncovering patterns and relationships in data, however, these tools can be challenging to learn, apply correctly and are also open to abuse.

  • How Machine Learning can be used to analyse huge amounts of data, uncover patterns and relationships, and define a trading edge,
  • How Machine Learning tools can be abused and the common mistakes that traders make with Machine Learning,
  • Strategy validation techniques that best suit market data and 1 popular technique that shouldn’t be used,
  • How to approach the vast libraries of algorithms available today,
  • Why delaying the trading process can lead to opportunity cost and how to know when a model is ready for trading.

A Guided Tour of Machine Learning for Traders – Dr. Tucker Balch


Which algorithms really matter for investing? In his presentation, Professor Balch helps declutter the Machine Learning jungle. He introduces a few of the most important ML algorithms and shows how they can be applied to the challenges of trading.

Talk Overview 

  • Machine Learning: Big Picture
  • Decision Trees: Classification
  • Decision Trees: Regression
  • Decision Trees Example: Sentiment-based strategy
  • kNN: Classification
  • kNN: Regression • Reinforcement Learning

Slide: link

Market Timing, Big Data, and Machine Learning – Dr. Xiao Qiao


Return predictability has been a controversial topic in finance for a long time. We show there is substantial predictive power in combining forecasting variables. We apply correlation screening to combine twenty variables that have been proposed in the return predictability literature, and demonstrate forecasting power at a six-month horizon. We illustrate the economic significance of return predictability through a simulation which takes positions in SPY proportional to the model forecast.

The simulated strategy yields annual returns more than twice that of the buy-and-hold strategy, with a Sharpe ratio four times as large. This application of big data ideas to return predictability serves to shift the sentiment associated with market timing.

Slide: link


Trả lời

Mời bạn điền thông tin vào ô dưới đây hoặc kích vào một biểu tượng để đăng nhập: Logo

Bạn đang bình luận bằng tài khoản Đăng xuất /  Thay đổi )

Google+ photo

Bạn đang bình luận bằng tài khoản Google+ Đăng xuất /  Thay đổi )

Twitter picture

Bạn đang bình luận bằng tài khoản Twitter Đăng xuất /  Thay đổi )

Facebook photo

Bạn đang bình luận bằng tài khoản Facebook Đăng xuất /  Thay đổi )


Connecting to %s

%d bloggers like this: