About

I have been working off and on as a residential and commercial appraiser since 2001, becoming a designated member of the Appraisal Institute in 2006.   However, most of my work experience since graduating from the University of Oregon with a degree in Mathematics, has  been as a software engineer, with a few other occupations thrown in such as photography and statistics.  

My appraisal work is noteworthy because I use programming and statistics in every appraisal, particularly R language and MARS (aka “earth”) analysis to create price  models.  

Valuation is split into two parts:

1.  Valuation of the property based on measured attributes such as GLA, Lot Size, Room Counts, Location, etc..

2. Valuation of the property with respect to the unmeasured or subjective features such as condition, quality, functional utility, aesthetics/design and view. The method for doing this is far more objective than with nearly all other appraisers, utilizing a scored residual approach for the sales comparables that is  completely objective and then ranking the subject property against the sales comparables based on comparison of the residual subject attributes against the ranked listing of the comparables.  This approach arguably reduces the error considerably,  estimated less than +/-2%  for over 80% of homes in most market areas in the SF Bay Area, — under the assumption that the MLS data itself is accurate.