QUESTION: Thank you for taking the time to participate in the Q & A session. Before we start in detail. Would you like to share with everyone who may not know, who you are, where you live and what is your current occupation?
Aaron: My name is Aaron Miles, I’m a Data Scientist based in Columbus, Ohio. I live with my wife, two sons (5 and 8), dog, and rabbit.
QUESTION: You have a very unique background as a data scientist now getting involved in retail trading. I sense you will be incorporating both skill sets in the near future. Can you explain?
Aaron: The two skills I think apply best to trading are distributional modeling and automation. With distributional modeling (as detailed in some blog posts here), I can get insight into the expected range and skewness of a security, and trade in such a way to profit from it.
Another skill I hope to use is automation. There are thousands of securities traded on the major exchanges, with the software skills I’ve learned throughout my career, I can set up a system where stocks are automatically scanned, profitable trades identified, and those trades executed, all at the touch of a button. In a nutshell, I hope to be able to identify and execute profitable trades in an automated way.
QUESTION: What personal or professional failure/setback have you experienced in your personal life or professional career that has set you up for later success?
Aaron: One of the most disheartening things I hear in my career is ‘you build exactly what we told you to, but we can’t use it for [reasons]’. In these cases, my team and I built what we thought was an awesome system with exact detail to client specifications, but some barrier that nobody anticipated prevents the system from actually being used.
In trading, I’ve built models that estimate a security’s movement from one day’s close to the next, and they are fairly accurate for a trading model. However, this is no guarantee of profitability at all, you have to think about after and before hours movement, entry points, etc. in order to actually profit. In my career, it’s easy to just think about a model’s accuracy, but that’s only a small part of the picture.
QUESTION: How would you describe what makes a successful Trader as it relates to making correct decisions based on statistical math probabilities?
Aaron: Two things, seemingly contradictory. First is a trust in your model. If you don’t trust your model, or the probabilities you’re working with, you shouldn’t be using them at all. One paradigm I’ve heard is that a good model is better than no model, and no model is better than a bad model.
Second is an almost paranoid attitude to your model being wrong. Every model is based on underlying assumptions, and you need to have a critical view of whether those underlying assumptions are still being met. The 2008 crash is a great example. Renaissance Technologies is the gold standard for quantitative funds; they definitely produce good models. But when some of the underlying assumptions of financial markets stopped being valid, Jim Simons had them halt trading, according to Greg Zuckerman’s book. Always be thinking critically about your model, this will both allow you to quickly identify and act on changes, and also build your own belief in your model when things are going smoothly.
QUESTION: What do you feel is the role in having a balance all around life in your day to day activities. What hobbies do you enjoy?
Aaron: You have to have balance. Your brain simply cannot focus on one thing all the time. My main hobbies are golf, exercise and weightlifting, video games, and wrestling with my kids. Many of my breakthroughs in work have happened when I’m doing these activities and not mainly thinking about work.
QUESTION: As a data scientist, what attracted you into the world of Trading?
Aaron: Conceptually, one thing I love about my data science in general is when I’m able to systematically predict the future, or the outcome of an event. There’s nowhere that’s more profitable than in trading. It’s also REALLY hard, and has kind of become my white whale. Whenever I learn a new technology for my job, I think about how it can be applied in the trading world.
How did it become my white whale? A few events kind of pushed me toward where I’m at now. First, we met with a wealth management firm for work, and I was fascinated by the data they were using, how they were making predictions, and what they were doing in general. Shortly after this, I went to the ODSC conference in Boston and attended some financial modeling workshops. Sitting in these workshops, I remember having the thought that ‘I can do this.’ and I got work that night.
Another pivotal event was the Big Data Bowl held by the NFL. The goal of this competition was to predict an entire distribution of possible rushing outcomes. This wasn’t a type of modeling I had done before, and thankfully I work with some awesome people who taught me a lot about it. I immediately saw an application to trading and have been working on a distributional modeling system ever since.
QUESTION: What good or bad advice do you hear often in life?
Aaron: I might just be an ornery contrarian, but I kind of cringe at how often I hear ‘Focus on process over results’ when it comes to trading. I get the idea of this quote, a trader who struck gold on a single YOLO trade is likely a worse trader than the person who is systematically extracting a half a percent from the market, even though p/l might way otherwise. But you always need to be focusing on whether or not your process is a good or bad process. If you’re following your process and consistently losing, then you probably need to ditch the process and find a new one!
QUESTION: Where do you see yourself in 5 years?
Aaron: Still moving along in my data science career, and having trading be a consistent supplementary income. Other than that, wherever the wind blows me!
QUESTION: When you have lost your mojo or focus , what do you do to get back on track?
Aaron: Some kind of exercise, it helps me focus and slow down all the things going on in my mind. Most of the time when I’ve lost my focus it’s because I’m restless and I need to get that energy out. Yes, I’m kind of a 5 year old in that regard.
QUESTION: If you could have a big billboard with your favorite saying or message on it, what would it be?
Aaron: ’You have to play to win.’ I don’t know if trading will ever be as consistent a money maker as I want it to be, but I know it WON’T be if I just sit on the sidelines.
Question: Thank you for taking the time for our readers to get to know you. How can our readers contact or follow you on social media?
Aaron: I’m most active on twitter at @_aaronmiles. I’m also on LinkedIn. I try to regularly write at my blog, where I post about things I’m learning about probability, data science, finance, and fantasy sports.