John Roberson’s a well-experienced equities trader, who’s been able to harness the power of technology and leverage the skills of his team to maximize his trading operation. Since starting out in 2007, John has been involved in various trading groups, and today he’s with Kershner Trading Group (Texas).
Although John originated as a discretionary click trader, over the years, he begun to automate more and more. To the point that he now runs a portfolio of 80+ automated models, which do everything from news and earnings, to mean reversion and momentum, to arbitrage and micro structure…
Topics & Timestamps:
Please note: Exact times will vary depending on current ads.
- 02:55 – The reason John started trading, and the “tiny” edge that enabled him to become profitable (which also bought him time to uncover additional edges).
- 13:45 – John’s evolvement to automated trading, early difficulties, and his “pull the plug” moment when a malfunctioning stop-loss order rocked Chevron stock.
- 30:15 – Developing a toolkit for backtesting, the power of machine learning and most used algorithms: XG boost, ordinary least squares, K-means clustering.
- 48:55 – Running a portfolio of 80+ automated models, families of strategies, exploiting anomalies, suggestions for click traders who wish to automate.
- 57:00 – Obsessing over your edge, adding layers, and thinking “really small.”
Links & Resources
- @Stanley223 [Twitter]
- John Roberson [Quora]
- John Roberson [LinkedIn]
- Regulation NMS [Investopedia]
- Chevron: Malfunctioning Stop-Loss Order [Quora]
- HFT Rocks Chevron [Nanex]
- Market Wizards [Amazon]
- The Top Ten Algorithms in Data Mining [Amazon]
- Machine Learning with Random Forests & Decision Trees [Amazon]
- Advances in Financial Machine Learning [Amazon]
- Introduction to Machine Learning [Amazon]
- The Last Place on Earth [Amazon]