[During our last Institute for Innovation Development Financial Services Innovation Forum in NYC, we asked Andrew Dassori, Founding Partner and Chief Investment Officer of Wavelength Capital Management to share with us his experiences and personal journey in creating next generation portfolio management tools and strategies utilizing artificial intelligence (AI) and machine learning. His firm has built a unique AI investment platform to work with advisors and money managers offering an artificial intelligence overlay for portfolio managers and hedge funds, a mutual fund and a robo-advisor.]
Dassori: I am here to talk about my experience with artificial intelligence, its application to investing, and how I see it changing the future of investing.
My personal investment journey
I started investing at 18 when I opened an ETRADE account and a friend's father lent me his copy of Ben Graham's Intelligent Investor. Before placing my first trade, I read it cover to cover and tried to draw out all the investment logic it contained. This was not easy, but I kept at it and eventually had an actionable set of logic for my own investing.
Graham was brilliant and he had to be. Not only did he establish a set of investment principles that remain relevant today, he also maintained and processed them in his head to make investment decisions. This highlights a key difference between investing intelligently back then versus today - he invested without the advantage of a computer to maintain his logic and execute his trades. This advantage has grown from its roots in codified investment logic to more advanced predictive analytics and forms of artificial intelligence.
How do you keep up with the volume of data, speed, and new opportunities?
Investment processes like Graham's are powerful, as they've been tested continuously by time. What has changed most is the way they can be applied given an ever-increasing volume of data. There are now hundreds of thousands of securities in markets across the globe, and the speed at which we access data has increased meaningfully. The value of different data points also changes over time, so while still applicable, the opportunity to apply these concepts is an expanding and moving target. As a result, it has become more difficult to manage an investment process in your head. We are fortunate that the explosion in data has been met with an explosion in computing power which far surpasses the capacity of any human brain.
Much of this was already in motion when I began investing so I recognized a changing opportunity set from the start. Codifying logic like Graham's lets investors keep up with change and leads to artificial intelligence in its base form as a rules-based system.
How did you build your investment process?
In engineering something repeatable that can also adapt to new environments, there are many basic parallels to the process of building an engine. Our investment engine runs on data which it processes through rules. These rules are codified as parts in the engine which can be thought of as if-then statements. Each rule is tested for consistency over time and maintained using a computer. Just as an engine's capacity is measured in horsepower, the power of our investment process comes from the logic of many minds working together. As a result, this engine can quickly analyze huge amounts of data using far more logic than what is possible to keep in one's head.
The output of our investment engine is applied to our portfolio which adapts logically to changing conditions over time. As humans, we check the output to make sure it is consistent with the engine's design but prevent the common pitfalls of emotion through disciplined, systematic execution. This key advantage lets us to spend our time analyzing new pieces of logic and building processes to identify these automatically through machine learning.
How do you expect machine learning to affect the future of investing?
Machine learning is useful across many disciplines, and its application to investing continues to evolve. When you consider a rules-based system, each rule can be improved, and borrowing an idea from genetics this can happen through evolutionary computation. In this, sets of trading rules are processed through different market scenarios, and at the outset the strongest or most profitable survive. These combine with other surviving rules to create stronger, more profitable rules for the future. Increased computing power in turn increases the number of rules that can be tested across an ever-expanding set of data. These conditions are what make machine learning such a compelling tool for investing, and at the same time they are what make it harder for humans to compete.
What other forms of artificial intelligence could provide an investment edge?
Another area where artificial intelligence can be applied to investing is the process of deep learning, as machine learning often faces a challenge in feature extraction. To solve this, deep learning was initially used for facial and speech recognition, and is based on how we, as humans, recognize these. The process uses neural networks, effectively connecting neurons by mimicking the design of the human brain. While still in the early stages, deep learning can be applied to financial markets in seeking to recognize patterns in the prices of securities.
How can advisors remain competitive as the investment landscape changes?
Rules-based systems have been used for decades and change is inherent in financial markets. Much of the logic used in these systems has also been consistent since being examined by the likes of Benjamin Graham. As such, investment value remains fundamentally determined by future cash flows. The key difference is in how these principles are applied, whereby offering clients access to new approaches can create meaningful value.
We have built an advanced AI-based investment platform so that advisors can continue to compete and grow as technology changes the industry. Because we built a flexible system that can be applied in different ways and different markets, we can develop different services and products for money manager clients focusing it to their specific investment areas as an overlay tool. Our strength is that our investment signals are applicable to any investor or advisor. We can essentially plug this system into managed accounts or ETFs and plug into any platform, if anyone is looking for an enterprise solution. Our first AI-based product is a mutual fund, Wavelength Interest Rate Neutral Fund (MUTF:WAVLX), and now we are rolling out a robo-advisor, Longwave Advisor, our white-labeled, digital interface for advisors and their clients, that offers access to AI-driven return predictive signals and automated portfolio management so they are not left behind as others adapt.
Investment advisors remain a key component of the financial system and in this an engine for growth. Those who operate ahead of the curve and use new tools to deliver principled investment advice are positioned to make the greatest impact moving forward.
by Bill Hortz, Founder & Dean, Institute for Innovation Development
The Institute for Innovation Development is an educational and business development catalyst for growth-oriented financial advisors and financial services firms. We position our members with the necessary ongoing innovation resources and best practices to drive and facilitate their next-generation growth, differentiation and unique community engagement strategies. The institute was launched with the support and foresight of our founding sponsors - Pershing, Voya Financial, Ultimus Fund Solutions, Fidelity, and Charter Financial Publishing (publisher of Financial Advisor and Private Wealth magazines).
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.