EP 090: This quants’ approach to designing algo strategies—Michael Halls-Moore

Aaron Fifield Podcast 1 Comment


For this episode I’m joined by Michael Halls-Moore, who runs QuantStart.com—a site well-known by many algorithmic traders.

Prior to trading, Michael studied computational fluid dynamics and was the co-founder of a tech startup, before getting involved a small equity fund as a quant developer—where his key role was cleansing data.

Now, independently, Michael trades his own short-term algorithmic strategies, consults to hedge funds on machine learning and quant infrastructure, and also has a keen interest in space exploration.

We discussed a whole range of topics, including; the need for quality data, thinking about risk from a portfolio level, trading multiple automated strategies, the role of common sense in parameter optimization, learning to program, and more.



Lessons learned in this interview:

  • Michael shares his story as a startup founder prior to trading, how he got involved with an equity fund as a quant developer—with a sole focus on data cleansing.
  • Why Michael believes risk management models not only limit losses and drawdowns, but also give traders the ability to make traders more money.
  • Insight to how Michael thinks about risk from a portfolio level and trades a series of, fully-automated, short-term trend following and mean reversion strategies.
  • Michael discusses what could be the most difficult area of quantitative trading; parameter optimization, and when common sense beats raw numbers.
  • How Michael makes use of machine learning techniques, particularly cross-validation, and does artificial intelligence have a place in the future of trading?
  • What programming language is best for developing an algorithmic strategy, how to learn a language as a non-programmer, and where to turn when you get stuck.

Follow Michael on Twitter: @MHallsMoore

Links and resources mentioned:

  • [Articles] Time series analysis – Here you’ll find all articles Michael has written about time series analysis, and here’s the article mentioning satellite imagery.
  • Anaconda by Continuum Analytics – Michael recommends installing Anaconda to get started using Python, which includes many popular data science packages.
  • Quantocracy.com – This website is a link aggregator for quantitative trading articles and content from around the web—one of Michael’s top-rated resources.
  • StackOverflow.com – The ultimate online community for finding answers to programming questions and troubleshooting code error messages.
  • QuantStart.com – To learn more about Michael, visit QuantStart where he writes about all aspects of algorithmic trading from beginner to advanced subjects.
  • @MHallsMoore – Follow Michael on Twitter, and tweet him about this episode!

How to support this podcast:

  • For a quick and easy way to support this free podcast, please write an honest review in iTunes. It’ll take you two minutes, and it helps massively—thank you.