Appraisal Accuracy and Automated Valuation Models in Rural Areas

Abstract

Accurate and unbiased property value estimates are essential to credit risk management. Along with loan amount, they determine a mortgage's loan-to-value ratio, which captures the degree of homeowner equity and is a key determinant of borrower credit risk. For home purchases, lenders generally require an independent appraisal, which, in addition to a home's sales price, is used to calculate a value for the underlying collateral. A number of empirical studies have shown that property appraisals tend to be biased upwards, and over 90 percent of the time, either confirm or exceed the associated contract price. Our data suggest that appraisal bias is particularly pervasive in rural areas where over 25 percent of rural properties are appraised at more than five percent above contract price. Given this significant upward bias, we examine a host of alternate valuation techniques to more accurately estimate rural property values.

An Investigation Into the Use of “Hybrid” Adjustment Techniques in the Application of the Sales Comparison Method in Residential Valuation

“…These studies cover a number of issues, such as valuation accuracy (Aliyu, et al, 2018), uncertainty in valuation (French & Gabrielli, 2004;Kucharska-Stasiak, 2013), information asymmetry (Garmaise & Moskowitz, 2004;Firoozi,et al, 2006), valuation variances (Munshifwa, et al, 2016) and others. Thus, the reality is that, in practice, modifications are made to this basic model by practitioners as dictated by the type of property being valued, differences in the volume of market information available, technological advancements, increased client awareness, demands for more accuracy in the assessed values and generally changed dynamics in real estate markets (Addae-Dapaah & Ho, 1996;Aluko, et al, 2004;Appraisal Foundation, 2014;Bogin & Shui, 2020). The Appraisal Foundation (2014) also shows that sales comparison is the underpinning approach in AVMs, through the direct market model or comparable sales models.…”

“…Literature further points to the shortcomings of the basic sales comparison model, thus advocating for further advancements in its application. Bogin and Shui (2020), for instance, argued that property appraisals tend to be biased upwards. This upward biasness is more pronounced in rural areas where there is fewer sales evidence and houses are characterized by greater heterogeneity (Bogin&Shui,2020); a comparable situation to what is found in most real estate markets in the global South.…”

Metrics for evaluating the performance of machine learning based automated valuation models

“…In the literature, MSE and R 2 are used to evaluate AVM performance by Kok et al (2017), Masias et al (2016), Bogin and Shui (2018), and Yacim and Boshoff (2018)…”

Boosted Tree Ensembles for Artificial Intelligence Based Automated Valuation Models (AI-AVM)

“…While licensed appraisers have the obligations to provide unbiased opinions on housing values, they may also in some instances be succumbed to pressures of related parties to bias their appraised values. Appraisal bias that inflates values of residential real estate used as collaterals is not uncommon, where borrowers are motivated to use higher appraisal values to increase the leverage as measured by the loan-to-value ratio of mortgage loans (Agarwal, 2007;Agarwal et al, 2015Agarwal et al, , 2017Bogin & Shui, 2020). It is thus necessary to reduce human errors by strengthening the science in the appraisal process.…”