Regression for GLA Adjustment

[glenn walker]

I had a family member buy a new home in San Diego county in 2004 for $640k, hold it until 05/2016 (12 yrs) and resell for $610k. Not only did they lose $30k off the nominal sale price + their costs of sale, + the costs of the home improvements they added, but they also paid ~$1000/mo in additional mortgage payments than they would have paid in rents. So no, "now" is not always a good time to buy RE.

And if he had held until 2020 its $1.5 Million - Point is he is a bad buyer and a bad seller LMAO

 

[George Hatch]

Clearly, the most probable price is not a fixed benchmark, it moves relative to other sales as of the effective date.

However, the MV definition IS a fixed benchmark. We use the same defintion of MV for a 200k house as for a 2 million $ house. We use the same definition of MV in a down market as in a hot market.

If I were theoretically to appraise a same house for a refinance year after year for 5 years and other than updating the house stayed the same, if prices rose or fell I would opine a different OMV each year - year one 600k, year two 630k, year three, 700k, year four, 640k, year five 590 k ( lol the market tanked in year five and it was red hot in year three). But the MV definition was the same each year, and used to vet terms of sale for the comps as well as to create the HC terms of sale for my subject MV opinion. w

If they're making reference to what's typical and probable in the market then by definition we're getting into comparisons and context between these sales. Not comparisons to external benchmarks or even comparisons with what happened a couple years ago.

If I list this property today, how much can I expect to sell it for? If I want to buy that property today, how much can I expect to buy it for? Less than what anyone else is paying because they have FOMO fever and I don't? Or the same as them because I have no other alternatives in the market?

 

[Fernando]

I had a mortgage broker who sold his house too soon but still made a profit. If he held on to the property, he would have made more profit today.

 

[J Grant]

If they're making reference to what's typical and probable in the market then by definition we're getting into comparisons and context between these sales. Not comparisons to external benchmarks.

I never said most probable price was an external benchmark, idk, maybe you imagined somehow I said that?

The most probable price is in relation to other prices as of the eff date.

The MV definition however does not change and remains the same in every assignment (usually)- and that defines the terms of sale and financing conditions we vet prices by. Why is this even a thing, this is appraisal 101 or should be...

 

[George Hatch]

I never said most probable price was an external benchmark, idk, maybe you imagined somehow I said that?

The most probable price is in relation to other prices as of the eff date.

The MV definition however does not change and remains the same in every assignment (usually)- and that defines the terms of sale and financing conditions we vet prices by. Why is this even a thing, this is appraisal 101 or should be...

Finally. It's based on what these sales are actually doing in the here and now. Nothing else.

 

[sputnam]

"Credibility" has to be the lowest bar among all standards ever devised by humans, and is essentially untested in the appraisal world because no one will pay to compare the result to reality. Anyone can write a credible report completely lacking in veracity, and since the practical application of "credible" has become, "can we make the loan?", there really is no standard in that. You simply cannot reconcile the "most probable" aspect in the definition of market value with the lack of complete lack of value in "credible" results as "supported", our other non-existent standard of quality. The clear result is that the bulk of lending valuations will soon be conducted outside the toothless/useless/worthless auspices of USPAP and the existing licensing and regulatory bureaucracy.

Maybe.... but an appraisal is not a fact that is found. It's an opinion that is developed and, hopefully, supported. It's not math. It uses math but, there is no one right answer as to what the subject property is worth. Given that truth, 'credibility' is a high bar.

 

[J Grant]

Maybe.... but an appraisal is not a fact that is found. It's an opinion that is developed and, hopefully, supported. It's not math. It uses math but, there is no one right answer as to what the subject property is worth. Given that truth, 'credibility' is a high bar.

I agree, and would add that while there is no one "right" perfect numerical answer, the range of credibility lies strongly around one numerical $ opinion rahther than others.

The strong point of appraising is there is an established set of standards to measure against and the data itself serves as a check - if one appraiser uses comps very similr to the subject needing fewwer adjustments and another appraiser uses comps in superior areas and that are much higher quality with huge adjustments ( or no adjustments when needed) then the second appraisal is of lower credibility - credibility means worthy of belief -

Review appraisals exist as another check

Compare that to an AVM or any machine valuation, where there really is no viable check on its credibility or conclusions. The only check is a confidence score for an AVM, or if a regression/statistic spits out an absurd result - the problem is some believe any math result is perfect because in their mind, math is never "wrong" - yet it's application can be fatally wrong, even if "accurate.".


Two atomic bombs and two apples both are represented by the number 2 in math. So while math can be accurate, in and of itself it can be fatally flawed. Both have the digital number 2 , but would any sane person compare 2 apples to 2 bombs? Math does, when it lacks human context and judgment .And AI reaches a programing limit if one argues AI can be programed to replicate human thought.

 

[Terrel L. Shields]

If he held on to the property, he would have made more profit today.

If I had bought a 100 shares of Walmart in 1980, I'd be a millionaire today... so?... 200, 400, 800, 1600, 3200, 6400, 12,800, 25,600, 51,200 x $157

The most probable price is in relation to other prices as of the eff date.

Sales, by definition, are historical. Therefore, none of the historical data can define price as of the effective date perfectly. Close perhaps, but not perfectly. That's why costs, listings, and even contracts are to be analyzed as well.

 

[NJAPRAZE]

i really feel bad saying this, but with my big city work & lots of close comps, sometimes the only adjustment i have to make is GLA. and it's usually because i have to bracket the GLA
& even then not a big $ amount..
certainly the data set you have does affect the amount of adjustments. easy to get a good grade on your report. in my young days i did do suburban & rural suburban, so i did pay my dues to have my current privileged appraiser life.

in phila or pittsburgh?

 

[Mejappz]

Maybe.... but an appraisal is not a fact that is found. It's an opinion that is developed and, hopefully, supported. It's not math. It uses math but, there is no one right answer as to what the subject property is worth. Given that truth, 'credibility' is a high bar.

While I agree that an appraisal is an opinion, regression can test how credible and accurate that opinion is. In the analysis below, all 46 sales were within 5% of their actual price. How well do you think it predicts the subject? Look at record 47. It is estimated that the subject would sell for 570K. It sold for 565K. There is a 95% chance the subject value is between $541,000 and $598,000.