Trunk-Branch Repeat Sales Index for Property Valuation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing property valuation methods face challenges in accurately predicting property values due to systematic and idiosyncratic errors in transaction data, particularly in handling multiple transactions and addressing biases in refinance transactions, which lead to aggregation and transaction type biases, and volatility in house price indices.

Innovation Solution

The implementation of a Trunk-Branch Repeat Sales Index (TB-RTI) that controls for systematic biases by using trusted purchase transactions for large areas and adjusting non-purchase transactions, along with data cleaning to mitigate idiosyncratic errors and Multiple-Transaction Based Property Valuation (MTV) that uses a weighted combination of multiple transactions to predict property values, incorporating recency and transaction type weighting.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If refinance transaction data is used to increase data coverage for HPI estimation, then the quantity of transaction data increases, but transaction type bias and volatility are introduced

Engineering Contradiction:

Improvequantity of transaction dataVSAvoidtransaction type bias

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The patent segments the HPI estimation process into two distinct components: a trunk HPI estimated from purchase transactions at a broad geographic level, and branch HPIs estimated from refinance transactions at local levels. This segmentation allows each component to serve its specific purpose while mitigating the biases of individual data types.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The trunk HPI acts as an intermediary that mediates between the biased refinance transaction data and the final local HPI estimates. By using the trunk HPI to adjust refinance transactions, the patent eliminates transaction type bias while preserving the increased data coverage provided by refinance data.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Quantity of substance

If large geographic areas are defined to ensure sufficient transaction data, then data sufficiency is improved, but aggregation bias increases

Engineering Contradiction:

Improvesufficiency of transaction dataVSAvoidaggregation bias

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent implements a two-level geographic segmentation: a broad trunk geographic area sufficient for reliable HPI estimation, and smaller branch geographic areas that capture local market heterogeneity. This allows sufficient data aggregation at the trunk level while enabling precise local measurements at the branch level.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by allowing different geographic areas to have different HPI estimates (branch HPIs) based on their specific local market characteristics. Each local market can have its own HPI trajectory, capturing neighborhood-specific price dynamics that would be obscured in a large aggregated area.

Inventive Principle:

Principle #3Local quality

3Measurement precision

If multiple prior transactions are used for mark-to-market valuation, then valuation accuracy is improved, but selection complexity and volatility increase

Engineering Contradiction:

Improvevaluation accuracyVSAvoidselection complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the parameter of transaction selection from a static single-transaction approach to a dynamic multi-transaction approach with time-varying weights. The weighting scheme evolves based on the number of available transactions, automatically adjusting the contribution of each transaction to the valuation.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements feedback by using the count of available prior transactions as a determinant for weighting. The system monitors the number of transactions and adjusts weights accordingly, providing feedback-based adaptation that reduces volatility while maintaining accuracy.

Inventive Principle:

Principle #23Feedback

4Ease of operation

If a single prior transaction is used for mark-to-market valuation, then simplicity is maintained, but valuation accuracy decreases due to idiosyncratic errors

Engineering Contradiction:

Improvesimplicity of valuationVSAvoidvaluation accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent merges multiple prior transactions into a single composite valuation estimate by combining them with appropriate weights. This merging process preserves the simplicity of a single valuation output while incorporating information from multiple transactions to reduce the impact of idiosyncratic errors in any single transaction.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS7765125B1Trunk branch repeated transaction index for property valuation

Publication Date: 2010.07.27 MAE FANNIE

AI summary

A repeated transaction index (RTI) for estimation of a home price index (HPI) that controls for systematic bias. Purchase transaction records are used to estimate an HPI for areas large enough to have sufficient transaction data. Bias is removed from non-purchase transaction records using the HPI. All available data is used to estimate a localized HPI that pertains to smaller geographical areas (e.g., zip codes, neighborhoods, etc.) after correction of the biases in the non-purchase transaction records.