Reserve order information can be a powerful tool to create alpha—even in its aggregated summary form. Recent Exegy research has shown that end-of-day data from one of our Signum predictive trading signals, Liquidity Lamp, can be used with machine learning to deliver four times better returns with 10 times less risk than following the S&P 500.
The development of this alpha strategy, which we call Diesel, is part of Exegy’s ongoing effort to highlight the breadth of uses for its Liquidity Lamp signal, which identifies reserve (iceberg) order volume on US exchanges. Using Liquidity Lamp, firms can cut through the noise of market data by focusing on and following the activity of the most informed investors, a practice known as “alpha cloning.”
Signum’s Liquidity Lamp Summary, an aggregation of a trading day’s reserve order activity, represents a rich data set that can be enhanced with machine learning models such as the one Exegy developed to create our Diesel alpha cloning strategy. To spur ideas for other innovative uses of this data that could assist in your strategies, please review our whitepaper by signing up here.
Liquidity Lamp Summary and the Power of Alpha Cloning
Liquidity Lamp, part of Signum’s predictive market data suite, analyzes patterns of orders to detect in real time which ones are being placed by large institutional investors attempting to hide reserve orders. While this information is valuable in the moment to assist in liquidity seeking and execution strategies, the data the signal produces over time has other marketable uses.
Liquidity Lamp Summary delivers an end-of-day aggregation of reserve orders on a per-stock, per-market basis. For each listed security, it summarizes trading activity that uses reserve order types, showing volume, notional value, and volume-weighted average price.
This information enables firms to see where institutional investors are moving without having to wait weeks or months for required federal filings.
Alpha cloning, sometimes called “coattail” or “copycat” investing, is an effective way of following the behavior of the top asset managers without incurring their fees.
Adding machine learning to the mix enables firms to move from pattern detection to actual prediction—using recent market activity to forecast the coming day’s reserve order volume and respond ahead of the market. It’s an approach that has generated impressive results.
Diesel: A Machine Learning Driven Strategy
As outlined in our whitepaper, we arrived at a model that delivered significant value: Relative to the S&P 500 our model achieved four times better returns with a whopping 10 times less risk.
Applying this model to data stretching back to 2018, the trend line is stable and provides remarkable downside protection, even during market meltdowns such as the March COVID-related drop.
Diesel is only one of many uses that Exegy foresees for its Liquidity Lamp Summary data. We’re working on more—and we anticipate that firms will mine it for further insights. With automated delivery in machine-readable format, it’s ideally configured for use in systematic strategies.
Want to learn more details about our example Alpha model and explore Liquidity Lamp Summary further? Read the whitepaper, request demo data, or talk to our team about how your strategies might be enhanced with predictive market data.
Want to learn more about Diesel? Get a guided tour of our alpha dashboard that shows how our strategies generate unique alpha.