Here’s what ML and NLP powered in capital markets in 2022

Innovation is the name of the game, especially in the capital markets. And two prominent tools behind innovative developments are machine learning (ML) and natural language processing (NLP).

The use of AI is rapidly growing in all facets of life, including trading. It’s growing so fast, in fact, that regulators including the Bank of England and the Dutch AFM have raised concerns about the possibility that machine learning trading algorithms could ‘learn’ to collude. Quants are also questioning AI’s model validation powers.

Still, it cannot be denied that more and more, vendors, banks, asset managers, and fintechs are incorporating these subsets of artificial intelligence into platforms, workflows, and systems aimed at simplifying tasks and unlocking new insights. Here’s a look at some key ML and NLP advancements we saw in 2022 (click on links for the full article):

Citi: The investment bank has developed a model that forecasts the shape of implied volatility surfaces for its single stock options book under multiple scenarios, each relating to different spot moves after an announcement.

The model uses gradient boosting, a type of machine learning that iteratively improves a tree-based classification model, training on spot price data and changes in the shape of the implied volatility surface. Citi’s quants retrain the model weekly.

Man Group: The London-based hedge fund has revamped its data science platform, Arctic—and now called Arctic Native. Man Group has a 400-strong business-wide technology unit. Within it sits the Alpha Technology team, which Gary Collier leads, comprising 175 engineers and five business managers. The team set out to rewrite Arctic from the ground up—this time in C++. Collier says it took four years of “intense effort” to complete the version that exists today.

“Imagine a Microsoft Excel spreadsheet with 100,000 columns and 10 million rows. Now, imagine 1,000 such spreadsheets and trying to build a trading model based on them,” said Collier, head of Alpha Technology at Man Group.

He continued: “We thought we could do better ourselves and process tick data in volume across a high-performance computing cluster. That was the genesis of Arctic.”

FactSet: The data and trading tools provider is blending ML and NLP algorithms to unravel patterns in the data and predict future events. Its ML models can determine which data features have the most impact on a certain outcome, and in what combination it is likely to occur, using “massive amounts of time series financial data”. FactSet’s predictive signals can detect, for example, secondary equity offerings, the candidates that are likely to issue follow-on offerings, what companies are likely to issue new corporate bonds, and even activism.

FactSet’s developers are also employing techniques like word embedding, an NLP technique that can simplify neural networks and improve sentiment analysis. Word embedding has been injected into services such as Named Entity Recognition (NER), which identifies companies, people, locations, health conditions, drug names, numbers, monetary values, and dates from unstructured or semi-structured documents. NER also allows users to link any document with other FactSet content sets, such as historical prices or fundamental data.

Bloomberg: This year, Bloomberg’s Compliance Solutions unit, which includes Bloomberg Vault, made a strategic investment in London-based Insightful Technology. To help better monitor new communication channels—such as Microsoft Teams, Zoom, or WhatsApp—Bloomberg will integrate Insightful’s AI-driven Soteria compliance and surveillance system.

S&P Global: In 2018, S&P bought ML and NLP specialist Kensho for $550 million. Since then the startup’s AI tools have been integrated throughout S&P. For example, one tool that’s been rolled out is called Scribe, which is based on automated speech recognition technology, and uses AI algorithms to process raw audio recordings of people talking—for example, an earnings call—and produce transcriptions.

Kensho has also launched tools called Nerd and Extract. Nerd is an NLP capability for identifying contextual clues, abbreviations, and potential redactions. Users can match unstructured information from regulatory filings, earnings calls transcripts, press releases and the like with the structured information in the massive S&P Capital IQ database. Extract is a machine learning tool trained to carve out hidden data from sources of unstructured data like PDFs.

Kensho is currently developing another tool, Classify, a classification system that labels documents based on topics and themes. The aim is to provide a tool that an analyst could use to organize documents based on themes that interested them, even if those topics change day-to-day.

Barclays: Working on the bleeding edge of machine learning, the investment bank’s quantum computing experts are developing a method to speed up machine learning models in finance. The work has been tested on time series data such as stock prices. Until now, a lack of suitable hardware has slowed progress in real-world applications of quantum computing, but the new method converts quantum outputs into binary states and could prove a viable alternative to quantum memory.

Infima: The startup, which was spun out of a project at Stanford University, claims its approach delivers a 10x improvement in the accuracy of predictions around MBS valuations and changes. The vendor has more than 30 bulge-bracket sell-side and buy-side firms testing its system, and just raised $5 million in seed funding from Radical Ventures, an AI-focused private equity firm, along with other investors including Franklin Templeton and ThirdStream Partners.

Infima utilizes Amazon Web Services and Microsoft Azure to pull in “billions of single data points” and then combines that data with its proprietary non-parametric deep-learning model to find hidden patterns in borrower behavior.

BMLL Technologies: The London-based analytics vendor is providing granular futures data to New York University’s Mathematics in Finance program. The academics want to run computations on market activity to understand the behaviors and impacts of trades, both buying and selling. The team has previously focused on the equities market in their research, using artificial intelligence and deep learning to run computations, and it now wants to apply the same research methods to futures.

Swarm Technology: The UK-based hedge fund runs a $95 million systematic trading strategy based on the way ants interact. The trading program, known as Swarm XVI, consists of individual algorithms that share information with each other, rather like an ant colony. In the 40 months from inception through to October 2022, Swarm XVI has generated a return of 57.29%. The S&P 500 over the same period has gained 30.55%.

Mosaic Smart Data: The European Space Agency hopes that banks and asset managers can use machine-learning algorithms that were initially designed to monitor instruments aboard deep-space satellites to manage risk better and find previously unforeseen trading signals more efficiently. The intergovernmental organization, which is run by 22 member states that include the United Kingdom, Spain, Germany, and France, where it is headquartered, is partnering with Mosaic Smart Data, a real-time data analytics firm that focuses on fixed income, currencies, and commodities trading. Mosaic is aiming to solve for the so-called fat-finger error, where the wrong key is mistakenly pressed on a computer to input data.

Claira: The startup is applying AI tools to read financial documentation. Claira was co-founded by Eric Chang, a former trader at Goldman Sachs, BlackRock and AQR, who went on to a role in AI product development at Morgan Stanley; Joseph Squeri, who was a CIO at Citadel and Barclays; and Alex Schumacher, an expert in NLP and a subset of NLP called natural language understanding (NLU). The three worked together at Exos Financial, a digital bank founded by former Credit Suisse CEO Brady Dougan in 2018.

Claira aims to save financial professionals time and effort by providing a pre-trained tool that gives them a better understanding of documents. Claira’s specialty is the kinds of documents that are heavily negotiated and bespoke, like CLOs, though it can also be used to parse 10k filings and the like. The platform leverages NLP and recent developments in NLU to convert financial contracts into business logic—customized algorithms that define how a business operates.

NN Investment Partners (Goldman Sachs): The asset manager, which oversees more than $330 billion of assets (but which has since been acquired by Goldman Sachs Asset Management), uses NLP to measure stock moves at times when announcements about dividends or buybacks occurred.

A breakdown using NLP of earnings call transcripts during the early days of the Covid-19 pandemic, for example, showed a 12 standard-deviation fall in how much executives were referring to earnings, and a seven standard-deviation spike in talk about social issues. The exercise revealed a similar increase in firms talking about company culture on earnings calls, a nearly 20% uplift.

SS&C Technologies: In March, SS&C closed its $1.6 billion acquisition of Blue Prism, a company that specializes in robotic process automation. While there are those in financial services who feel that RPA is a “Band-Aid” rather than a root-cause remedy for manual processes, Michael Megaw, managing director of business process automation at SS&C, said there’s a misconception about so-called digital workers.

“You can point digital workers in the direction of the work that needs to be done—RPA models don’t restrict you in that regard—and it’s typically less expensive to maintain,” he said. “I disagree with the sentiment that RPA is legacy technology. The capability of these digital workers is only getting better; we think that RPA is a component of getting our customers scale, and we aim to continue investing in the product and enhancing its capabilities.”

In its finance vertical, SS&C will first look to point Blue Prism’s digital workers at its suite of compliance tools, specifically for KYC/AML needs. It is also pairing Blue Prism with its own business process optimization tool, SS&C Chorus.