12:12 PM
Connect Directly

Arialytics Unveils Predictive Analytics Platform

Aimed at asset managers and hedge funds, the platform enables investors to test their investment ideas without getting bogged down in data sets and modeling.

Arialytics announced the release of an enterprise analytics program specifically built for asset managers, fund-of-funds, hedge funds and institutional investors to streamline the investment discovery and modeling process.

The platform called Aria combines Big Data with predictive analytics for the investment discovery process. It allows fund managers and analysts to create and test their own investment ideas without getting bogged down in the data cleansing and modeling process.

“Aria empowers both the fund manager and the analyst alike to answer many research questions, very rapidly,” said David Marra, founder and CEO of Arialytics, who contends it takes away the time consuming work of aggregating and cleansing data.

“What Aria does is it abstracts you away form all the statistics, away from building data sets and predictor variables and optimizing parameters,” explained Marra, who spent his career at Boston Consulting Group where he worked in financial services and pharmaceuticals. Users can bring their own data or Aria can provide it for them. “We call it accelerated data cleansing and accelerated predictor creation,” said Marra.

With traditional investment research, firms may spend 80% of their time on tedious technical activities, leaving only 20% or less for insight driven work, states the firms' release.

The raw data can come from the financial markets, or firms can use unstructured data like social media from various aggregators. Also, an analyst who covers retail stocks can also focus on specific data, such as retail data.

“It’s not asset class dependent. It’s any kind of data, any kind of asset class. It’s discover and prediction,” he said.

According to Marra, the most value is in the modeling because modeling takes up a lot of time. “You’ve got to achieve a statistical tool of which there are hundreds and you have to optimize the parameters,” he explained. "Aria objectively looks at millions of parameter tools and combinations. It basically automates the modeling process and it does that by optimizing algorithmic and parameter selection," he said.

Marra used the analogy of turning on the ignition in a car and it automatically starts, whereas the alternative would be to set the electrical system and turn all the levers before starting the car.

“The bottom line is you can realize investable solutions in a matter of days versus months,” said Marra.

The product is geared to firms like fund-of funds that don’t have technical shops to answer questions. “It really plays into these firms that don’t have any army of PhDs. “It was built so that people can focus on an idea, and not the technical part of modeling,” he said. Investors can ask different questions: For instance, someone wants to know how well it could predict the euro one day ahead, as well as two days ahead or third days head, etc. “It allows someone to drive massive research agendas by allowing you to answer all those questions in a few days,” said Marra.

While the models are built off of historical data, Aria monitors the market conditions in real time in order to figure out an optimal solution at every point in time.

After working on the R&D for three and one-half years, Arialytics launched the platform today. Marra developed the methodology while working in pharmaceuticals for BCG. He worked on projects related to drug pipeline approvals, in which only a fraction of the drugs are going to get approval. “It’s a predictive analytic problem,” said Marra, who developed a methodology for how to predict which drugs had high likelihood to be approved and be profitable.

Outlining some of the ways in which asset managers could use the platform, Marra said long only equity funds could use Aria to analyze their currency exposures and hedging costs. “Hedging costs eat into their alpha and their costs,” said Marra.

A lot of firms deal in very illiquid assets, that don’t trade very frequently, so it’s difficult to understand what the true value is. “Aria can look at a much larger data sets and add in a predictor set to get much better pricing for illiquid things,” he said.

In the fund-of-funds industry, where people are building portfolios by selecting other funds, they want to know the risk they are taking on. Someone can use Aria to build very sophisticated benchmarks to compare one fund vs. another to make sure it meets its mandate, he said.

Because people are so used to researching investment ideas the same way, the firm bundles a data scientist with the software to stimulate them to use the new tool. According to Marra, the tool can result in a five-to-ten fold gain in productivity.

Ivy is Editor-at-Large for Advanced Trading and Wall Street & Technology. Ivy is responsible for writing in-depth feature articles, daily blogs and news articles with a focus on automated trading in the capital markets. As an industry expert, Ivy has reported on a myriad ... View Full Bio

Register for Wall Street & Technology Newsletters
7 Unusual Behaviors That Indicate Security Breaches
7 Unusual Behaviors That Indicate Security Breaches
Breaches create outliers. Identifying anomalous activity can help keep firms in compliance and out of the headlines.