AI: The Best Shot at Extracting Alpha

Stockholm (HedgeNordic) – Artificial intelligence and machine learning technologies have been developing at a fast pace over the past few years, with possibilities soaring in tandem with the ever-increasing volume of data, advancements in computing capability and data storage solutions. Asset managers may need the genuine adoption of artificial intelligence and machine learning to stay ahead of their game.

“The complexity of financial markets is such that no one individual or team at a hedge fund or elsewhere can absorb all the information and extract all the alpha you can get unless you use technology,” asserts Daniel Broby (pictured), Director of the Centre for Financial Regulation and Innovation in the United Kingdom. “The advantage of artificial intelligence is that you are not just using the technology to process and classify available data, you are using it to make predictions and come to assessments faster than the rest of the market.”

“The complexity of financial markets is such that no one individual or team at a hedge fund or elsewhere can absorb all the information and extract all the alpha you can get unless you use technology.”

The applications and benefits of artificial intelligence, however, can vary greatly from one asset manager to another. “What AI can bring to the table very much depends on the strategy and what you decide you want your AI to actually deliver for you,” explains Broby, who has produced a number of papers on the use of artificial intelligence in banking and fund management. “The specific strategy of a hedge fund manager is very relevant to how you program artificial intelligence and what you want to get out of it. Whoever has the best processing power, the best understanding of the data and the best ability to translate that into a model has the best shot at extracting alpha right now.”

Applications of AI for Hedge Funds

Artificial intelligence has grown its presence across the asset management industry, having the ability to transform many facets of the industry. Different forms of artificial intelligence such as machine learning and natural language processing are being used and can be used across the industry to improve portfolio management, trading, and risk management practices, among many other things. “You can identify how artificial intelligence can benefit things and gain a competitive advantage,” points out Broby, who enjoyed a successful career in the Danish asset management industry prior to joining academia. “Over time, the advantage will disappear because everyone will be employing artificial intelligence, but right now, there is a gap and that gap is there to be exploited.”

Broby recently co-wrote a paper on how AI applications in satellite imagery and data can improve sustainable finance outcomes, with these applications having the potential to create new sources of alpha for hedge fund managers as well. “Satellites now have large amounts of data that can either be used to oversee or identify trends,” Broby discusses his paper. “Using satellites, you can get better monitoring than you can from a self-reported regime, which is where we are right now,” he points out. “If you want to look at manufacturing output or want to measure CO2 emissions from factories, AI can do that very precisely right down to individual factory level,” he provides one application of AI using data from satellite imagery.

“The insights you can get from all this information, if processed with big data, allow you to have an informational advantage over the market,” reckons Broby. “But bear in mind that this is an immense amount of data, you can’t put it into a spreadsheet in excel,” he cautions. “You have to put it in a flat file and use a programming language to work out the insights.” The sub-set of alternative data stemming from satellite imagery “can give a hedge fund an edge,” reckons Broby. “Using a mosaic approach, you are processing information better than anyone else, thereby giving you the potential to achieve alpha.”

“Hedge funds can also benefit by using natural language processing. It is far quicker to look at sentiment now on things like social media such Twitter or stock exchange feeds by identifying textual words using what we call a “corpus” and looking for trends in those.”

“Hedge funds can also benefit by using natural language processing,” asserts Broby. “It is far quicker to look at sentiment now on things like social media such Twitter or stock exchange feeds by identifying textual words using what we call a “corpus” and looking for trends in those,” he elaborates. As a subset of alternative data sets, NLP can be used to conduct sentiment analysis, determine information relevance, among other things. “The technology to do it is there,” says Broby.

Other of Applications of AI

Machine learning algorithms, as a subset of artificial intelligence, are used in a wide variety of applications both inside the asset management industry and beyond. Broby, for instance, relied on machine learning to develop a patented tool for valuing private companies in real-time. The tool combines publicly available data on private companies from Companies House in the United Kingdom to create a sample of ratios, which are then processed to estimate “efficiently-priced” valuations for private firms using stock market data. “Whether or not fund managers agree that the stock market efficiently prices companies, there is evidence that in aggregate is actually the case,” asserts Broby.

“Taking our sample of ratios from private markets that we have processed using programming techniques, we then use machine learning to assign a more accurate valuation to those based on current market prices,” he explains. The valuations assigned by the patented valuation technology “reflect the sentiment that is going on in the market,” says Broby. “The valuations of private companies are not dynamic and are not priced in real-time. Obviously, the valuations we are creating are.”

“The other area which is ripe for development and improvement due to artificial intelligence is also the interaction between markets,” reckons Broby. “There are 14 odd different stock markets in the United States, various commodity markets in the world, the futures and options markets,” he highlights. “Using technology, markets will move closer and closer together in the future, and that will create more opportunities for new innovative strategies.”

AI Can Kill Alpha Too

Most discussions about artificial intelligence revolve around the advantages its applications can bring to the asset management industry. But can AI stop some hedge fund strategies from generating alpha? Broby certainly thinks so. He co-wrote a paper on position timing and distributed ledger technology, which has the potential to wipe off the profits enjoyed by high-frequency traders in the hedge fund space. “Markets are going to move to the internet and you don’t have a central marketplace on the internet, you have distributed market places,” explains Broby.

“The age of getting ahead of an order may go. It is not just who is going to win, but also who is going to be put out of business because of information advantage disappearing.”

“As investors send messages between these distributed ledgers, as you send blockchain transaction orders, these have to be cleared effectively to meet buys and sells,” he elaborates. “As you send them in nanoseconds rather than having a central market place aggregate in seconds, these orders effectively don’t match up, so you need to bundle the orders together.” In collaboration with the National Physical Laboratory, Broby “worked out the effective bundling time to minimize the effect of frontrunning.” By reducing the ability of high-frequency traders to front-run other investors, the group could lose billions of U.S. dollars in profits. “The age of getting ahead of an order may go,” reckons Broby. “It is not just who is going to win, but also who is going to be put out of business because of information advantage disappearing.”

Nordics and Generational Leaps

According to Broby, the Nordic asset management industry may be better positioned to benefit from the increasing adaptation of artificial intelligence relative to other regions. “The graduates who enter the finance industry in the Nordic region have more experience and knowledge of how to program than do graduates in other markets, and therefore, have an information processing advantage,” reckons Broby, who launched Denmark’s first regulated and listed daily pricing hedge fund. “Likewise, in my experience, there is a greater focus on wanting to experiment and implement the outcome of those experiments in the Nordics,” he continues. “The desire to innovate, plus the skillset to implement, is an advantage.”

“The graduates who enter the finance industry in the Nordic region have more experience and knowledge of how to program than do graduates in other markets, and therefore, have an information processing advantage. The desire to innovate, plus the skillset to implement, is an advantage.”

Whereas the new generation of tech-savvy investors and asset managers are increasingly deploying artificial intelligence, senior executives at larger asset managers may lack a foundational understanding of artificial intelligence. “Senior managers at asset managers are a generation away from what is going on,” asserts Broby. “What I learned when I was at business school and what is currently being taught is totally different, the theory of finance has now been added with programming, data analytics, distributed ledgers and so forth. It is a generational leap,” he continues. “Senior managers of asset management Companies – not hedge funds because hedge funds are a bit more cutting edge – really need to either listen more or adapt their processes.”

 

This article featured in HedgeNordic’s report “Technology and Hedge Funds.”