Automated Property Valuation Adjustment via Transaction Feedback

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Automated Property Valuation Models (APVMs) face inaccuracies due to idiosyncratic errors and systematic biases, such as time lag, price tier effect, and transformation bias, leading to inaccurate property value predictions that can increase credit risk and affect business operations.

Innovation Solution

A valuation adjustment scheme (VAS) is implemented using purchase transaction data to derive location-specific adjustment factors, specifically addressing time lag and price tier biases, and correcting transformation bias through data-driven methods, thereby improving the accuracy of property value predictions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated property valuation models are used to predict property values, then productivity is improved through automated valuation, but measurement precision deteriorates due to systematic biases and idiosyncratic errors

Engineering Contradiction:

Improvevaluation efficiencyVSAvoidproperty value prediction accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system implements feedback by comparing automated valuation model predictions against actual purchase transaction data, calculating adjustment factors that capture systematic biases. These adjustment factors are then fed back into the valuation process to correct future predictions, creating a continuous improvement loop that maintains both automation and accuracy

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The invention changes the parameters of the valuation system by introducing location-specific and time-specific adjustment factors that modify the raw model predictions. These parameters account for systematic biases such as time lag effects and price tier effects, transforming the output from biased predictions to corrected valuations while preserving the automated efficiency

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If purchase transaction data is used to derive adjustment factors, then measurement precision is improved by reducing systematic biases, but device complexity increases due to additional data processing requirements

Engineering Contradiction:

Improveprediction accuracyVSAvoiddata processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system applies local quality by creating location-specific adjustment factors for different geographical areas rather than using a single universal adjustment. This allows the system to account for local market conditions and systematic biases unique to each location, improving precision without requiring overly complex global models

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The valuation adjustment process is segmented into distinct components: base model prediction, systematic bias identification through transaction data comparison, adjustment factor derivation, and final corrected valuation. This segmentation makes the complex data processing more manageable and systematic

Inventive Principle:

Principle #1Segmentation

3Loss of time

If valuation databases are updated monthly, then loss of time is reduced through regular updates, but measurement precision deteriorates due to time lag bias in the data

Engineering Contradiction:

Improvedata freshnessVSAvoidvaluation accuracy

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The system performs preliminary action by deriving adjustment factors from the most recent purchase transaction data available before finalizing the valuation. This allows the model to anticipate and correct for time lag biases before they fully impact the valuation, maintaining both timeliness and accuracy

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The invention applies preliminary anti-action by identifying and counteracting time lag bias through adjustment factors derived from recent transaction data. The system proactively corrects for the inherent delay in monthly database updates by incorporating the latest available market information into the valuation adjustment process

Inventive Principle:

Principle #9Preliminary anti-action

Data Source

PatentUS7788186B1Method and system for automated property valuation adjustment

Publication Date: 2010.08.31 FANNIE MAE

AI summary

A method and system are provided removing systematic bias from property value predictions obtained using automated property valuation models. The systematic bias is removed by deriving monthly adjustment factors that are location-specific. The monthly adjustment factors are obtained by comparing newly obtained purchase transaction data to model-generated valuations of properties in a base valuation database. The newly obtained purchase transactions may be obtained from newly obtained data on recently completed purchase transactions and loan application data on future purchase transactions. The monthly adjustment factors are applied to the property valuations in the base valuation database to obtain adjusted property values having reduced bias and thus improved accuracy.