From Model to Market: Reflections from FILS on AI, Pricing, and Relative Value in Fixed Income
By Timothy Stevens, CFA, Chief Product Officer, SOLVE
One of my biggest takeaways from FILS was that the fixed income conversation around AI has become much more practical. The industry is no longer debating whether advanced data science has a role in trading and portfolio management. The focus has shifted to where it belongs: how firms can apply better data, better models, and better workflows to make faster, more confident decisions in live markets.
That came through clearly in the panels and conversations around pricing. As the volume of available data continues to grow — from eval prices and dealer runs to customer bids, TRACE prints, and live indications — firms are wrestling with how to turn all of that information into something usable. Some firms are building internal pricing systems. Others are relying on vendors. Many are doing both. But the underlying point is the same: pricing has become the foundation for automation, scalability, and better investment decision-making.
1. Quality Pricing is the starting point for effective investment analysis
Analytics, AI and machine learning are only useful when they are grounded in high-quality data. In fixed income, that is not a small challenge. Markets are fragmented, liquidity varies widely, and the best view of value often comes from combining many different signals. The goal is not simply to collect more data. It is to establish a reliable view of where a bond is likely to trade.
That is where predictive pricing becomes so important. At SOLVE, we use AI to generate live signals from market data and predict where a bond will trade, with models designed to minimize the difference between the predicted level and the actual trade outcome. Clients can use that insight in multiple ways: as a trading signal, as an input into their own internal pricing models, or as a guardrail around automated and AI-driven workflows. One large buy-side firm found that SOLVE’s AI-generated pricing would have produced greater profitability on 70% of trades compared with its internal model.
2. Explainability matters as much as innovation
Another theme that stood out at FILS was governance. As firms adopt AI more broadly, they need to understand not only what a model is suggesting, but why they should trust it. In practice, explainability comes from clearly defining success, testing models rigorously, understanding prediction error, and giving users confidence around the outputs.
I discussed this in greater depth in SOLVE’s on-demand webinar, From Model to Market: Turning AI-Driven Fixed Income Insights into Action, where we explored how better signals, faster decisions, and more accountable investment strategies depend on the quality and transparency of the data and models behind them.
This is especially important in fixed income, where decisions often need to be made quickly but also defended clearly. Traders and portfolio managers need tools that support human judgment, not black-box recommendations. Better data and better models should help users ask sharper questions, compare alternatives more effectively, and act with more confidence.
3. Relative value leverages pricing to support optimal decisions and becomes actionable when paired with live bids and offers
The most exciting application of these themes is Relative Value Analysis™. Traders and portfolio managers are overwhelmed with thousands of bonds, bids, offers, and market signals. It is difficult to know which opportunity is truly attractive, which bond is cheap or rich versus true peers, and which market segments deserve more attention.
With SOLVE Px’s predictive pricing and SOLVE Quotes’ live pre-trade data, relative value can move beyond static comparisons. Users can evaluate a bond against another bond, a bond against a customized cohort of similar bonds, or one cohort against another. That matters because fixed income decisions are rarely made in isolation. Traders need to know whether a bond is attractive versus comparable alternatives and whether that opportunity is actually tradeable. Portfolio managers need to understand where value may be building across sectors, ratings, maturities, and other market segments.
That was my biggest impression coming out of FILS: the firms that create an edge use AI and analytics to bring together trusted data, explainable models, and workflows that help teams move from insight to action. Pricing is the foundation. Relative value is where that foundation becomes decision support. Live bids and offers turn all of this into actionable insight.
Related Resources
About SOLVE
SOLVE is the leading market data platform provider for fixed-income securities, trusted by sophisticated buy-side and sell-side firms worldwide. Founded in 2011, SOLVE leverages its AI-driven technology and deep industry expertise to offer unparalleled transparency into markets, reduce risk, and save hundreds of hours across front-office workflows. With the largest real-time datasets for Securitized Products, Municipal Bonds, Corporate Bonds, Syndicated Bank Loans, Convertible Bonds, CDS, and Private Credit, SOLVE empowers clients to transform the way they bring new securities to market, trade on secondary markets, and value highly illiquid securities. Headquartered in Connecticut, with offices across the globe, SOLVE is the definitive source for market pricing in fixed-income markets.
