#United States #Irvine #Real Estate #ATTOM #AVM

ATTOM Unveils Next-Generation AI-Powered Automated Valuation Model

In a significant milestone for the real estate industry, ATTOM has introduced an innovative automated valuation model (AVM) that harnesses the power of artificial intelligence. This model marks a departure from traditional valuation methods, which have often grappled with limitations, especially in a challenging housing market characterized by low transaction volumes.

Breaking Away from Tradition

Traditional AVMs typically depend on recent sales data to generate property valuations. This approach can lead to inaccuracies, particularly in fluctuating markets filled with fewer transactions. With more than three decades of transactional data, ATTOM's new AI-driven model relies on a comprehensive understanding of property histories, allowing it to provide reliable assessments that transcends conventional methods.

ATTOM’s AVM is designed to address the specific challenges faced by lenders, insurers, and investors in low-liquidity environments. Relying on a vast database that includes time-adjusted historical transactions, this groundbreaking model enables accurate property-level evaluations that are crucial for underwriting, portfolio analysis, and risk management.

Enhanced Accuracy Through AI

Through extensive out-of-sample testing over the last ten years, ATTOM's AVM has shown consistently high reliability, achieving a median absolute percentage error of just 2.9%. Notably, over 80% of the valuations generated fall within 10% of the actual sales prices. This impressive accuracy stems from the model's ability to learn over time, analyzing property characteristics, historical pricing trends, and hyperlocal insights.

Rob Barber, the CEO of ATTOM, stated, “This is not simply an incremental enhancement; it’s a complete redesign of the valuation architecture using cutting-edge AI technology. We have progressed from static, comp-based methods to a dynamic system capable of adapting in real time to shifts in market behaviors.”

Aaron Wagner, the Vice President of Data Science at ATTOM, elaborated on the model’s capabilities: “We evaluate how neighborhoods have evolved over decades, using this temporal insight to convert historical sales data into precise current valuations. This dynamic approach offers a richer learning foundation for our valuation system, ensuring heightened accuracy in markets where traditional methods often falter.”

Confidence Scores and Enterprise Applications

An integral feature of ATTOM's AVM is its confidence score, which indicates the reliability of each valuation. This transparency allows organizations to automate decisions with increased confidence.

The AVM is built to serve various enterprise applications in mortgage, insurance, investment, and proptech fields. It offers flexible delivery options such as APIs, bulk data delivery, and cloud platforms like Snowflake and Databricks.

In addition, ATTOM provides a comprehensive technical white paper detailing the model's methodology, confidence metrics, and accuracy benchmarks, which will further assist stakeholders in leveraging this powerful tool for their needs.

About ATTOM

As a leading provider of property data, ATTOM boasts one of the most reliable property data assets in the United States, covering approximately 160 million properties that represent 99% of the population. Their multi-sourced real estate data encompasses various aspects, from property taxes and deeds to environmental risks and neighborhood dynamics. This rigorously validated information supports advanced analytics, helping users in diverse industries automate their property intelligence efforts.

For more information, media representatives can reach out to Megan Hunt at ATTOM. Explore how this cutting-edge AVM can reshape the dynamics of real estate valuation in an increasingly data-driven world.