RiskSpan Unveils Credit Model 7.1: Revolutionizing NonQM Analytics Within Its Platform

RiskSpan Unveils Credit Model 7.1



RiskSpan, a prominent player in the realm of data, modeling, and analytics for structured finance, has announced the launch of its latest innovation, Credit Model 7.1. This newly developed model is specifically tailored to meet the unique demands of the NonQM (Non-Qualified Mortgage) market, and it is integrated within the RiskSpan Platform.

As the NonQM sector has experienced significant expansion, with a notable surge in RMBS (Residential Mortgage-Backed Securities) issuance—almost doubling year-on-year—there has been an increasing need for advanced analytical tools that can keep pace. A report from Morningstar DBRS indicates that Non-QM RMBS issuance reached a staggering $20.9 billion in the third quarter of 2025, marking a remarkable 97% increase from the previous year.

The Capabilities of Credit Model 7.1


Credit Model 7.1 enhances the capabilities of RiskSpan’s platform by providing a complete credit model specifically designed for NonQM collateral. Notably, this model evaluates transition states by taking into account various documentation types, including Bank Statement, DSCR (Debt Service Coverage Ratio), Full Doc, and others. It incorporates ten key factors at both the borrower and loan level—such as FICO scores, loan-to-value ratios, debt-to-income ratios, and the purpose of each loan—along with three overarching macroeconomic drivers. This comprehensive approach is grounded in robust training data, utilizing around $87 billion in Outstanding Principal Balance (UPB) from approximately 226,000 NonQM loans between January 2018 and August 2025.

Additionally, the new Credit Model comes equipped with advanced AI-powered tools that include tape cracking and collateral analysis capabilities, alongside API access that allows clients to seamlessly integrate model outputs into their existing workflows. An upcoming user-facing backtesting dashboard will further enhance the functionality of the model.

Market Need for Tailored Solutions


“NonQM borrower behavior varies significantly according to documentation types, and standard credit frameworks don’t capture that variability,” noted Divas Sanwal, RiskSpan’s Head of Modeling. “Credit Model 7.1 has been meticulously crafted from the ground up, specifically for NonQM collateral. Our model is rigorously validated and offers risk teams and auditors the transparency they require to operate confidently.”

Before the introduction of Credit Model 7.1, the NonQM analytics landscape was primarily serviced by third-party models that were not specifically designed for the intricacies of NonQM loans. This gap in the market has left investors and issuers struggling to find suitable analytical tools that can keep up with the rapid evolution of the NonQM sector.

RiskSpan's introduction of a purpose-built prepayment and credit model, coupled with proprietary historical data and workflow tools, sets it apart from competitors. Active now, the Credit Model 7.1 is readily available to RiskSpan platform clients, with further deployment phases and integration plans set for the near future.

About RiskSpan


RiskSpan is dedicated to providing a single, comprehensive analytics solution for both public and private credit investors in loan and structured finance. Their innovative offerings empower clients to make quicker, more precise trading and portfolio management decisions—all while easing reporting burdens without the hassle of managing multiple vendors or internal systems. With advanced cash flow and valuation models, automated data pipelines, and quality controls, RiskSpan is prepared to meet the complex demands of institutional investors, dealers, and risk managers.

Learn more about RiskSpan and its groundbreaking solutions at www.riskspan.com.

Topics Financial Services & Investing)

【About Using Articles】

You can freely use the title and article content by linking to the page where the article is posted.
※ Images cannot be used.

【About Links】

Links are free to use.