Residual Constraint Approach (RCA): Framework & Protocol

William Bert Craytor

doi:10.67330/kgebxf46

作者

William Bert Craytor Author https://orcid.org/0000-0003-2219-9156

DOI：

https://doi.org/10.67330/kgebxf46

關鍵字詞：

不動產估價、住宅估價、殘差約束法、 MARS、多元自適應迴歸樣條、成交案例比較法、殘差分析、潛變數、自動估價

摘要

傳統估價方法的一個重大侷限在於其依賴過時的手工技術，例如配對比較分析（matched pair analysis）。這類方法對不動產特徵的價值貢獻視而不見：傳統的銷售網格（sales grid）沒有價值貢獻一欄，只有調整一欄。而事實上，價值貢獻正是建構數學約束的必要基礎——這些約束能夠防止傳統估價師的高估與低估。

殘差約束法（Residual Constraint Approach, RCA）透過整合一個多階段估價流程來推進傳統的成交案例比較法（Sales Comparison Approach, SCA）：該流程運用多元自適應迴歸樣條（Multivariate Adaptive Regression Splines, MARS），並對特徵價值貢獻施加嚴格的數學約束，尤其針對未被測量的特徵——諸如美學、設計、狀況與品質等潛變數——其價值在傳統估價中通常受制於估價師的主觀判斷與偏誤。

RCA 中的數學約束依賴於對 MARS 迴歸的專業運用，以估計可測量不動產特徵對成交價格的價值貢獻。然而，這一估計通常只能解釋實際成交價格的約 80%，不包括設計、功能效用、狀況和品質等主觀評估的特徵。在舊金山灣區（San Francisco Bay Area），代表這些主觀成分的殘差平均約占不動產總價值的 20%。80/20 這一數字反映的是作者在舊金山灣區典型資料條件下的經驗，不應被解讀為跨市場不變的常數。在可實現 R² 較低的市場中，本框架的應對之道是識別並測量未被測量的價值驅動因素——例如沿海社區中與海洋的距離、山坡社區中的海拔高度，以及當 MLS 資料未提供時從外部來源收集的類似變數——而不是接受這一侷限並報告更寬的誤差區間。雖然在其他情境下 MARS 殘差通常被視為估計誤差，但在不動產估價中，殘差是潛變數價值的間接度量。儘管這初看似乎不切實際，但已有證明表明：只要各描述性成分的貢獻之和等於殘差，殘差就可以被有意義地分解為這些成分——且不影響最終的估價結果。

這一方法論源自作者二十年的實證應用經驗，透過在 R 與 Python 環境中對 MARS 的迭代實作逐步發展而成。

下載

尚未有下載數據。

參考文獻

Ambrose, Brent W., Conklin, James N., Coulson, N. Edward, Diop, Moussa, and Lopez, Luis A. (2025). "Do Appraiser and Borrower Race Affect Mortgage Collateral Valuation?." Review of Finance. doi: 10.1093/rof/rfaf046.

Craytor, William Bert (2026). "The Doctrine–Practice Gap in Real Estate Appraisal: A Structured Account of Functions, Boundaries, and Tensions." Valuation Engineer Journal 1.1.

Friedman, Jerome H. (1991). "Multivariate Adaptive Regression Splines." The Annals of Statistics 19.1, pp. 1–141.

Hastie, Trevor, Tibshirani, Robert, and Friedman, Jerome H. (2017). The Elements of Statistical Learning: Data Mining, Inference, and Prediction. New York, NY: Springer.

James, Gareth, Witten, Daniela, Hastie, Trevor, and Tibshirani, Robert (2013). An Introduction to Statistical Learning: With Applications in R. New York: Springer.

Krivorotov, George and LaCour-Little, Michael (2020). "AVM versus Appraisal-Based Underwriting in Refinance Mortgages: The Trade-off Between Noise and Bias." SSRN Electronic Journal. doi: 10.2139/ssrn.3603327.

Kuhn, Max and Johnson, Kjell (2016). Applied Predictive Modeling. New York: Springer.

Milborrow, Stephen (2024). Earth: Multivariate Adaptive Regression Splines. https://CRAN.R-project.org/package=earth.

Milborrow, Stephen (2024). Notes on the Earth Package. http://www.milbo.org/doc/earth-notes.pdf.

Milborrow, Stephen (2024). Variance Models in Earth. http://www.milbo.org/doc/earth-varmod.pdf.

Perry, Andre M., Rothwell, Jonathan, and Harshbarger, David (2018). The Devaluation of Assets in Black Neighborhoods. Brookings Institution.

Perry, Andre M., Rothwell, Jonathan, Williamson, J., and Steinbaum, M. (2024). How Racial Bias in Appraisals Affects the Devaluation of Homes in Majority-Black Neighborhoods. Brookings Institution.

Property Appraisal and Valuation Equity Task Force (2022). Action Plan to Advance Property Appraisal and Valuation Equity. U.S. Department of Housing and Urban Development.

Tzioumis, Konstantinos (2016). "Appraisers and Valuation Bias: An Empirical Analysis." Real Estate Economics 45.3, pp. 679–712. doi: 10.1111/1540-6229.12133.

Williamson, Jake and Palim, Mark (2022). Appraising the Appraisal: A Closer Look at Divergent Appraisal Values for Black and White Borrowers Refinancing Their Home. Fannie Mae.

Zhang, Wengang (2020). MARS Applications in Geotechnical Engineering Systems: Multi-Dimension with Big Data. Singapore: Springer.