Mike Kennedy
Elite Member
- Joined
- Sep 28, 2003
- Professional Status
- Certified Residential Appraiser
- State
- New York
The Premier Online Community for Real Estate Appraisers!
-
Welcome to AppraisersForum.com, the premier online
community for the discussion of real estate appraisal. Register a free account to be able to post and unlock additional forums and features.
Appraisal Statistics
- Thread starter RCA
- Start date
- Status
DWiley
Elite Member
- Joined
- Apr 4, 2007
- Professional Status
- Certified Residential Appraiser
- State
- Tennessee
The adjustment grid on a 1004 is nothing more than the application of a statistical model, and we have been doing that for many decades now. So, what alternative do you suggest?Elliott said: ↑
The use of statistic modeling in real estate appraisal is appropriate when buyers and sellers start behaving with scientific consistency.
Nuff Said.
J Grant
Elite Member
- Joined
- Dec 9, 2003
- Professional Status
- Certified Residential Appraiser
- State
- Florida
Imo the adjustment grid is not a statistical model, it is rather, (after financing/market conditions adjusted for ) the most probable $ premium or $ deduction a borrower would make for features positive and negative using the best substitute comps , a small # of properties but highly qualified, vs statistics a large data of properties but poorly qualified .
The SC a models in grid form how the typically motivated buyer would decide a price among the most similar competitive properties. (and how long seller would have to exposure property to market to get that price ) ...
DWiley
Elite Member
- Joined
- Apr 4, 2007
- Professional Status
- Certified Residential Appraiser
- State
- Tennessee
Well, that really isn't a matter of opinion. It is not the same thing as using regression to analyze hundreds or thousands of sales, but it is certainly a statistical model.
When you adjust prices based on the anticipated reactions of typical buyers for difference in features, what do you think that is if it isn't statistical modeling?
J Grant
Elite Member
- Joined
- Dec 9, 2003
- Professional Status
- Certified Residential Appraiser
- State
- Florida
Perhaps...the SC grid of applying adjustments to predict behavior can be called a statistical model...though I doubt a "sample" of 3-4 sales and a listing would be taken seriously as statistics...
If an appraisal is nothing more than statistical modeling, why isn't it called a statistical model (or AVM ) ? It is called an appraisal because of its own methodology..(and USPAP compliance)...or what else earns it the name and distinction of being its own product..
I look at it that an appraisal answers a marketing question, not a math question , though math/statics can serve to derive or suppot the answer )
A math question can only answer another math question.
Does the math compute, do the statistics "support" their own result....(they have to, that is how they are set up to run data and spit something out. The software can't consider a market insight or perspective on that data, and the other problem is the data is mostly not vetted.
Seems the best of both worlds now with stats and appraisals 2 different products that can be used as comparison or in conjunction ...until real AI comes along, at which point all of us will be unemployed so who will be able to buy (or pay rent) on the properties...likely far future is human conscienceless avatars with no physical bodies needed ( too expensive for earth to support our needy resource consuming bodies, and we will not be needed much for labor or production or even ideas, once AI can generate ideas) , therefore no need for houses for bodies to live in or sleep or cook in...bye bye appraisals and AMC's too lol...at most a nostalgic online game for bored avators to play.
If an appraisal is nothing more than statistical modeling, why isn't it called a statistical model (or AVM ) ? It is called an appraisal because of its own methodology..(and USPAP compliance)...or what else earns it the name and distinction of being its own product..
I look at it that an appraisal answers a marketing question, not a math question , though math/statics can serve to derive or suppot the answer )
A math question can only answer another math question.
Does the math compute, do the statistics "support" their own result....(they have to, that is how they are set up to run data and spit something out. The software can't consider a market insight or perspective on that data, and the other problem is the data is mostly not vetted.
Seems the best of both worlds now with stats and appraisals 2 different products that can be used as comparison or in conjunction ...until real AI comes along, at which point all of us will be unemployed so who will be able to buy (or pay rent) on the properties...likely far future is human conscienceless avatars with no physical bodies needed ( too expensive for earth to support our needy resource consuming bodies, and we will not be needed much for labor or production or even ideas, once AI can generate ideas) , therefore no need for houses for bodies to live in or sleep or cook in...bye bye appraisals and AMC's too lol...at most a nostalgic online game for bored avators to play.
Last edited:
DWiley
Elite Member
- Joined
- Apr 4, 2007
- Professional Status
- Certified Residential Appraiser
- State
- Tennessee
Yes. Hence, the great irony (in my mind at least) of some of the posts in this thread, which are critical of the application of mathematically accepted statistical methods while,at the same time, inherently defending the use of very small sample sets (three to six comps) refined with adjustments that, truth be told, are all too commonly supported only by PFA analysis. To me, it seems as some inhabitants of glass-based dwellings are tossing about some mighty big geologic orbs. I urge caution.
DWiley
Elite Member
- Joined
- Apr 4, 2007
- Professional Status
- Certified Residential Appraiser
- State
- Tennessee
Pulled from....
If you selected an appraisal report at random, and asked the appraiser to provide actual support, using a recognized method or technique, for the GLA adjustment rate that he/she had applied, how many could produce that (without having to back into it now that the report was completed)?
J Grant
Elite Member
- Joined
- Dec 9, 2003
- Professional Status
- Certified Residential Appraiser
- State
- Florida
Yes. Hence, the great irony (in my mind at least) of some of the posts in this thread, which are critical of the application of mathematically accepted statistical methods while,at the same time, inherently defending the use of very small sample sets (three to six comps) refined with adjustments that, truth be told, are all too commonly supported only by PFA analysis. To me, it seems as some inhabitants of glass-based dwellings are tossing about some mighty big geologic orbs. I urge caution.
They are different methodologies..calling 3-5 comps a "Sample set" of statistics is a false equivalence...
The appraisal comps are not a smaller, weaker sample set of statistics.They serve their own purpose, as a small, but very strong set of human vetted and chosen sales/listings meant to replicate the most likely alternatives for a subject buyer would consider.
Who is the most important runners out of a field of many...the winners, the finalists, the number one two and three who get medals and a couple of others who almost timed for medal...the 1,2, 3 or more comps. The rest of the field ( statistical noise) nobody remembers after the race....they were there but in the bg and don't count at the end. A buyer is not going to consider 100 houses and pick any one of them, the buyer will narrow down their choices and goal is to find the "one" . Similar to marriage (for better or worse, ) people might date a number, then narrow it down to a few serious relationship "finalists" then close the deal by marrying just one. Since bigamy is not legal or accepted most places...some buyers own vacation homes./second investment properties but most are looking just for one, or at least one at a time.
Joe Flacco
Elite Member
- Joined
- Jul 31, 2013
- Professional Status
- Certified Residential Appraiser
- State
- Maryland
Yes. Hence, the great irony (in my mind at least) of some of the posts in this thread, which are critical of the application of mathematically accepted statistical methods while,at the same time, inherently defending the use of very small sample sets (three to six comps) refined with adjustments that, truth be told, are all too commonly supported only by PFA analysis. To me, it seems as some inhabitants of glass-based dwellings are tossing about some mighty big geologic orbs. I urge caution.
You need to be schooled by Leased Fee about statistics, regression, and real estate data.
- Status