"Sample" Appraisal: Subjective Value Containment Approach

[James Micozzi]

Somebody is trying to get at that effective date. Give up.

Give up what. As I already stated, based on all the information in your report as including the dates of the comps used in the report, a resonable person would likely agree with my opinion of the effective date. While the exact day of the month cannot be known it is irrelevant for this discussion.

 

[Digger88]

im not finding this form in total. couldnt this have been done on a 1004 in 2 hours?

 

[George Hatch]

It seems to me that if we want to test the value of something then the way to do it is to isolate the variable we want to test. If the question is about the methodology then why don't we run a paired comparison between the two using the same dataset? We could do the data qualification first so that each valuation model being tested would be using exactly the same data, qualified to the same degree however perfect or imperfect that may be.

 

[Michigan CG]

I don't know what you are talking about. Which comp?

It doesn't matter what comp. You made a $1,400 adjustment in an appraisal where the concluded value is in the $800,000 range.

 

[RCA]

To be honest, some redaction is necessary to make it look real. But

It doesn't matter what comp. You made a $1,400 adjustment in an appraisal where the concluded value is in the $800,000 range.

OK. What is wrong with that? Mars is that exact. Far more exact than a human. I can make these adjustments no problem. If you are using traditional methods, the argument can be made you are not capable of that kind of accuracy. I'm guessing that is kind of what you are talking about.

In other words, if MARS puts an adjustment out that amounts to $1,400 --- it has determined that it is significant. So, yes, I would use it.

 

[RCA]

To be honest, some redaction is necessary to make it look real. But


OK. What is wrong with that?

To be honest, some redaction is necessary to make it look real. But


OK. What is wrong with that?

It seems to me that if we want to test the value of something then the way to do it is to isolate the variable we want to test. If the question is about the methodology then why don't we run a paired comparison between the two using the same dataset? We could do the data qualification first so that each valuation model being tested would be using exactly the same data, qualified to the same degree however perfect or imperfect that may be.

??-

 

[RCA]

Give up what. As I already stated, based on all the information in your report as including the dates of the comps used in the report, a resonable person would likely agree with my opinion of the effective date. While the exact day of the month cannot be known it is irrelevant for this discussion.

Well of course, there is the possibility I did some global changes, search and replace in Acrobat. That never entered your mind I suppose.

 

[Red Flint]

... application of residual analysis to containing or limiting the subjective valuation, that I might respond to.

I continue to struggle with the ability to believe that there is a statistically relevant sample size to validate multivariate regression without using older data, which may not be reflective of current market forces. Or, have to include recent market data that does not correspond closely enough to the subject characteristics and hence introduces correlative dissonance. Then, after all of this modeling time, to try to build another model because the next subject has different attributes, especially for non-residential properties.

 

[RCA]

You made a $1,400 adjustment for an $800,000 house based on a regression with a 0.70 R2. That is less than 2/10th of one percent.

Oh yea, yea. BTW, the Stage I R2 is 0.91744 and Stage II is 0.9033 for the residual model, the two combined have an R2 of 0.99.2016%. But of course the residual R2 does not carry the same weight as the Stage I R2, because we have to fit the subject into that residual pattern --- and that is prone to at least some error. If you are off by 20% on the ranking of the subject, that then translates toa 0.10x0.20 = 0.02 or a 2% error. And that might just explain why I can often apply my model to a dozen or more comps and get them well withing 2 or even 1% - or even 0.5%.

You can't do that.

There is certainty with my results that far exceed those of other appraisers. Although, on occasion, I am in a neighborhood with poor data, and will have a low Stage I R2 - which effects Stage II and III.

 

[andrew81]

i mean this with the best intentions.

bert, you are too smart for us lowly residential appraisers.

seriously, you are putting way to much into this. stick to your software engineer work where your brain can be utilized for good.