As founder of Investment Science, Michael Kelly is deep into investing and financial know-how.
This is especially true when it comes to the data underneath it all.
Here’s how Michael envisions the future of the industry, including what he thinks it takes to succeed professionally even amidst black swan events.
Michael Kelly of Investment Science On Stock Algorithms and Consulting Without Compromise
How Investment Science Came to Be
The background of Michael’s company is really interesting.
“We originally started Investment Science all the way back in 2009 to be able to have stable returns through specific financial products and algorithms,” says Michael.
Essentially, it’s a concept that’s catered to different time frames and strategies using proprietary data elements. Now, his company has evolved more towards consulting, as a lot of his and his team’s skills developed from building out software.
Michael references the COVID-19 pandemic, which caused people to lose 15–30% from premier institutions in one fell swoop.
“We’re not going to say we could eliminate that, but we could potentially stabilize returns for all types of investors.”
Investment Science’s algorithmic product is still pre-launch and under wraps, but it essentially matches buyers and financial services. It’s not Betterment and it’s not Motif. Rather, the product is its own thing.
What’s It Take to Develop An Algorithm?
There are different types of algorithms.
Michael mentions execution algorithms, where a giant fund might hold a billion shares of Microsoft while measuring different metrics like average price. In that case, they may only execute a trade once the price crosses a particular average.
There are also statistical models, like statistical arbitrage, where algorithm creators look at pricing differentiation between financial products, time frames, and investing strategies.
For example, a foreign exchange spot transaction (FX spot) may provide you the opportunity to look at something as simple as GDP announcements, then ask yourself, “Historically, how do currencies react to that in a split second or over the course of the long term?”
The Algorithmic Balance Between Sentiment and Data, According to Michael Kelly
I’ve often thought that as the market evolves, we’ll eventually reach a point where news and public sentiment will dictate the moving of the markets. The algorithm that Investment Science is creating now is centered around sentiment as well as fiscal metrics.
Michael says the algorithm will be based on “really anything about a company, like a data warehouse. Any question you could have such as what time of day is the best time or with some of these sentiments, what things actually move together?”
There are always new data feeds, of course—like what happened with reddit and GameStop—that you’d have to monitor. Plus, it depends where your hardware is located, which is why a lot of banks keep theirs close to the exchanges in New York City (here, Michael mentions the concept of latency).
You cannot perfect these things (yes, there are companies that tend to consistently generate returns, but they’re not going to disclose how they do it without having you sign an NDA). But it is a free market with betters and sellers on either side of the equation. With that said, it’s possible to develop an adept algorithm that answers the biggest questions of the modern-day.
All in all, upwards of 85% of investments are done electronically. Instruments are popping up from every corner and the big players can’t push them all down.
What Michael Kelly Means by ‘Reverting to the Mean’
Despite the fact that he’s developing an algorithm for businesses and the market, Michael recognizes that there’s a bit of randomness to it.
“If you zoom back a bit, it’s kind of a diversion from the mean. Prices have an average and they will revert back to them at some time.”
Take a look at the coronavirus, for instance. Historically, it’s very similar to the Spanish flu. Different data points suggest that the Spanish flu lasted about two years, and there’s reason to believe the global pandemic we face today will follow the same trajectory.
Real estate is similar, in which companies panic and move out of the city without looking long-term. “That will revert to the mean, too,” Michael says.
And more recently, we saw a stock market anomaly that involved micro movers moving major markets. What happened there?
According to Michael, “That reddit forum turned into a mini fund of its own, just with the amount of participants and the amount of volume they could push.”
Some of these funds, Michael says, could also be considered black swan events in their own right. For business and retail investors, there’s no way to truly anticipate it, so the question remains: How do you resolve it?
“It’s something as simple as controlling your risk. Every dollar you invest, you have a max downside and you have potentially derivative instruments to offset that.”
Stop losses, for example, mean you can only lose so much. But sometimes you don’t have that liquidity, and things are great…until they’re not.
Consulting without Compromise
What’s the innovative stuff in the investment data industry? Michael knows.
“That’s what’s interesting about the company. We focus a lot on the business-to-business consulting advisory services. So we’re all about objective consulting without compromise, which really means you need us there a week—that’s okay. You need us there five years—that’s okay. We try not to have strategic ties to vendors, just giving you advice.”
Some of the latest trends involve graph databases and other databases that allow businesses to visualize unique scenarios.
Complicated coding language isn’t the driving factor anymore, either. Some organizations can leverage the technological hardware and keep things simple in terms of coding so they can get things done quicker and more efficiently.
From my perspective, I love coding as a means for digital marketing data. You can use that data for anything you want and build it into any interface and modeling you want. From an investment perspective, combining data science application with project and change management works great.
“With data science, it’s still evolving.”
On that note, will all businesses have some element of machine learning and AI in the next 5 years?
“Even the product we’re working on is supposed to make models drag and drop very simply and very affordable,” Michael says. “You don’t even need to know how to code. You just need to know what you want.”
That’s where everything is moving, even websites. Hardcore programmers themselves say that the next 5–10 years will see a general shift away from coding and toward configuration. Anyone with any background can work it for a low transaction cost.
“It’s more about the ideas for the product. It’s not even about the technology.”
The Investment Science Founder on Staying Ahead of the Game
We all see businesses pivoting.
Corporations are moving their headquarters out of major metropolitan areas and into smaller, more rural states with a lower cost of living. That’s not to say they won’t come back, but it’s a shift worth noticing.
On top of that, machine learning and analytics are only going to become more attuned. With that in mind, Michael recognizes just how important it is to maintain and develop soft skills to stay relevant.
For him, this means things like agile project management (“an iterative approach to managing software development projects,” according to Atlassian) and SAFe Scaled Agile Framework.
“The economy pivots every two years. As a consultant, I’m in between contracts every year, so I try to update my skill sets,” Michael says. “Stay on top of the trends and look at what companies are hiring, what position you want. Obtain that education, build those projects.”
Michael’s advice says it all. For more information about Investment Science, including tons of resources on data trends and investment information, head to www.investmentscy.com.
