Grid adjustment question

[J Grant]

50 years of knowledge superseded by an avm that says location has no special priority.

Your experience is absolutely wrong. One can drive 10 minutes in any direction my market and the same exact home can have 10% to 30% difference in value.

That is fundamental to appraising, and any method or model that does not account for it is not a viable method. In my area, properties 1/4 mile apart can vary widely in value (for a similar size and condition house or condo)

Location is fundamental to appraising because it matches what the market actually does, which is what an appraisal or any good valuation model, is supposed to do. The problem with mass data or using statistics that rely on a quantity of data is where the properties are not qualified or only in a very broad sense; thus, properties that are not similar to the subject start to influence results and can produce misleading results. The math is not misleading because math formulas correctly answer a mathematical question, but if the results don't hold true for the market, they are misleading.

Imo, regression and AVM's do not replace appraisals, they can be an alternate model and used as a check and balance for an appraisal, which indeed is what they are often used for.

 

[RCA]

That is fundamental to appraising, and any method or model that does not account for it is not a viable method. In my area, properties 1/4 mile apart can vary widely in value (for a similar size and condition house or condo)

Location is fundamental to appraising because it matches what the market actually does, which is what an appraisal or any good valuation model, is supposed to do. The problem with mass data or using statistics that rely on a quantity of data is where the properties are not qualified or only in a very broad sense; thus, properties that are not similar to the subject start to influence results and can produce misleading results. The math is not misleading because math formulas correctly answer a mathematical question, but if the results don't hold true for the market, they are misleading.

Imo, regression and AVM's do not replace appraisals, they can be an alternate model and used as a check and balance for an appraisal, which indeed is what they are often used for.

It is surprising how many appraisers, even MAIs believe this kind of nonsense. They invariably have never mastered advanced regression techniques like MARS, so they DO NOT know what they are talking about. They DON'T EVEN BEGIN TO KNOW - what they don't know.

Now, of course there are some neighborhoods where houses are very much alike - such as newer subdivisions - but even in these cases, we usually do not see pure linear relationships.

But honestly, I am wasting my time trying to talk to appraisers like you --- You have already read plenty of my posts, but still do not even begin to comprehend some relatively simple concepts. A very few on the forum however do appear to have a more advanced understanding.

But this is not the right forum. And, to be honest, I don't know that there is any "right" forum.

Eventually, this will change. It is just a matter of time. --- Although my guess is a lot of time.

In the background, there are always these people dreaming about becoming rich -- and they are the ones who are likely to move things forward, - although I think they will find it is not an easy path to riches.

 

[J Grant]

It is surprising how many appraisers, even MAIs believe this kind of nonsense. They invariably have never mastered advanced regression techniques like MARS, so they DO NOT know what they are talking about. They DON'T EVEN BEGIN TO KNOW - what they don't know.

Now, of course there are some neighborhoods where houses are very much alike - such as newer subdivisions - but even in these cases, we usually do not see pure linear relationships.

But honestly, I am wasting my time trying to talk to appraisers like you --- You have already read plenty of my posts, but still do not even begin to comprehend some relatively simple concepts. A very few on the forum however do appear to have a more advanced understanding.

But this is not the right forum. And, to be honest, I don't know that there is any "right" forum.

Eventually, this will change. It is just a matter of time. --- Although my guess is a lot of time.

In the background, there are always these people dreaming about becoming rich -- and they are the ones who are likely to move things forward, - although I think they will find it is not an easy path to riches.

Funny that you choose to call the fundamental of appraising, location and its infueolnce on properties, " nonsense" Everybody can have an opinion and this is your opinion but it does not mean it is valid. The same appraisal princles are used around the world and have been used scuscssfully for decades, on properties tht sell for 100k and on pfoprites that sell for ten million.

AVMs and other valuations, in fact, are computer-driven versions of appraisal, minus the human verification and qualification of properties. Some of them do a good job, others do not. Replacing appraisals with these alternate valusions is not necessarily better, it is cheaper and faster, and that is the main reason for its success. A nonappraiser or twelve-year-old can click on a Zillow valuation model and have it spit out data and a value in seconds. An appraiser is needed to develop an appraisal. Clearly, the two models serve different needs and purposes.

Y

 

[alebrewer]

OP - I agree with RCA in that, if using regression to estimate the GLA adjustment factor, and depending on the magnitude of the GLA differences, it can be janky. Next time you're faced with this, try a curvilinear regression to see if there's a better fit. It's been my experience that neighborhoods tend to have bookends WRT the 'preference' for optimal GLA. One neighborhood may express a range bewteen say 2,000' and 3,000'. Another at between 3,000' and 4,500'. The larger the homes, the wider the range typically. That said, any properties that fall outside the neighborhood 'preference' will typically look janky when you try to apply a consistent GLA adjustment factor. For homes 'in the range', the adjustment factor will be higher than homes above the range and lower than homes below the range - because of the law of diminishing returns. This can also help you confirm functional inadequacy or superadequacy.

Just don't try applying different GLA adjutment factors for F/F/F appraisals.... those underwriters typically aren't prepared to deal with that...

 

[RCA]

Funny that you choose to call the fundamental of appraising, location and its infueolnce on properties, " nonsense" Everybody can have an opinion and this is your opinion but it does not mean it is valid. The same appraisal princles are used around the world and have been used scuscssfully for decades, on properties tht sell for 100k and on pfoprites that sell for ten million.

AVMs and other valuations, in fact, are computer-driven versions of appraisal, minus the human verification and qualification of properties. Some of them do a good job, others do not. Replacing appraisals with these alternate valusions is not necessarily better, it is cheaper and faster, and that is the main reason for its success. A nonappraiser or twelve-year-old can click on a Zillow valuation model and have it spit out data and a value in seconds. An appraiser is needed to develop an appraisal. Clearly, the two models serve different needs and purposes.

Y

1. Why are you talking about AVMs?

2. Nonsense are statements like this: "Location is fundamental to appraising because it matches what the market actually does..." WTF? Location has many definitions or contexts. There is the location of a point on the map, of a specific property or house, of a block, of a neighborhood, of a market of a state, of a country. So, does the location of a house match the general market of a county? Does the location of a house match even the market for the house. It is rubbish. Or how about "Location is FUNDAMENTAL to appraising..." A BMW car in the SF Bay Area, probably has the same market value regardless of where it is. Location is important in appraising - just like many other property attributes such as GLA and Lot Size, elevation, view and so on. But two houses could be essentially in the same location and have much different values due to their other features. To say location is "fundamental" is kind of ridiculous. So are other features.

3. Again what does the Zillow vaiuation model have to do with anything. I certainly never use AVMs for appraising - although they may be used in discussion if better information is not available on some property under discussion. It does seem that you think that someone using regression is using an AVM. That is really far off base.

4. You talk about Mass Appraisal. But what does that have to do with anything? You apparently don't understand what Mass Appraisal is. You confuse it with the use of regression.

I could explain all of these things to you again and again. But, you just don't get it. You are that retrograde.

If I am wrong - give me a correct definition of Mass Appraisal and AVMs. See if you can do that.

But don't feel bad. A certain MAI that should be an expert on resdential appraisal told me a couple of days ago that because I was using MARS, I was doing Mass Appraisal. No! He was being a bit lightheaded.

 

[RCA]

I'll give you a partial definition of Mass Appraisal and be done with it:

1. Mass Appraisal is providing the estimated market value of a large number of homes in a given area.
2. Using MARS (or any other) regression to appraise a single property - is just that - appraising ONE property.

So, technically, what I do with comparables, which already have a sale price, is use the estimates and consequent residuals cranked out by the regression model to estimate the residual for the subject property. If I actually had to appraiser the 100+ comparables that go into the regression, then I would be going after their EXPECTED sale price or MARKET VALUE -which could be different than their sale price. I of course don't have any reason to do that and certainly not the time.

 

[Terrel L. Shields]

They DON'T EVEN BEGIN TO KNOW - what they don't know.

And some of us don't necessarily agree that you know what you think you know. After all, all the quants told us that there was no way the banks could fail in 2008 because contagion did not exist and MBS and CDOs were interlocking so perfectly they could not fall...in fact, they nearly took down the entire financial system.

Heuristic judgment suggests that the AI has to assume the same human foibles as buyers and sellers in pricing and making offers. Well, lemme see. AI is a construct of humans. And in examples, describing a citizen of Iceland, it came up with people of color, not white people. In fact, no matter how the question was phrased it included "people of color" in every grouping... imagine the KKK with black women, Asian men...etc. And you see the problem. Otherwise, you are blinded by the wisdom of the moment. It would be interesting to compare your formulaic methods against the actual sales of the properties you appraise.

 

[RCA]

OP - I agree with RCA in that, if using regression to estimate the GLA adjustment factor, and depending on the magnitude of the GLA differences, it can be janky. Next time you're faced with this, try a curvilinear regression to see if there's a better fit. It's been my experience that neighborhoods tend to have bookends WRT the 'preference' for optimal GLA. One neighborhood may express a range bewteen say 2,000' and 3,000'. Another at between 3,000' and 4,500'. The larger the homes, the wider the range typically. That said, any properties that fall outside the neighborhood 'preference' will typically look janky when you try to apply a consistent GLA adjustment factor. For homes 'in the range', the adjustment factor will be higher than homes above the range and lower than homes below the range - because of the law of diminishing returns. This can also help you confirm functional inadequacy or superadequacy.

Just don't try applying different GLA adjutment factors for F/F/F appraisals.... those underwriters typically aren't prepared to deal with that...

I do it.

1. Usually comps that go into the sales grid stand a good chance of falling in the same linear segment - but not always.
2. Sometimes I get a call from the AMC wanting an explanation for a adjustment for two houses that fall on two different linear segments, and I just explain it. The explanations are always in the appraisal to begin with - they just may need some further clarification. And I have been using MARS since 2004.
3. And note: It is simple and easier to calculate the value contributions for the subject and each comparable and then take the differences for the adjustment. But, you should be prepared to directly calculate the adjustments from the model. The math, even if it is simple high school algebra - gets a bit more convoluted than most people are used to. But heck you say: "This is just high school algebra." (Secret: People with a high SAT Math score can invariably deal with this.)

 

[Carnivore]

I found this using google

How to Build A House Price Prediction Model – Linear Regression Explained

Ever wondered how algorithms predict future house prices, stock market trends, or even your next movie preference? The answer lies in a fundamental yet powerful tool called linear regression. Don't be fooled by its seemingly simple equation – this ar...

www.freecodecamp.org

Then this from google


Linear regression models in real estate use mathematical equations to predict house prices based on property features. These models assume that the relationship between a dependent variable and one or more independent variables is linear

]https://towardsdatascience.com/pred...ne-learning-from-scratch-part-ii-47a0238aeac1[/URL]

https://www.researchgate.net/publication/321479223_Real_estate_value_prediction_using_multivariate_regression_models

So here is a Point that I want to make about the minimum educational path to Residential Appraiser State Licencing. This also includes the Supervisor/trainee training relationship that exist I assume in All States. it seems to me that maybe this should be part of the minimum educational requirement's for a trainee pre-education.

What do you think?

 

[J Grant]

1. Why are you talking about AVMs?

2. Nonsense are statements like this: "Location is fundamental to appraising because it matches what the market actually does..." WTF? Location has many definitions or contexts. There is the location of a point on the map, of a specific property or house, of a block, of a neighborhood, of a market of a state, of a country. So, does the location of a house match the general market of a county? Does the location of a house match even the market for the house. It is rubbish. Or how about "Location is FUNDAMENTAL to appraising..." A BMW car in the SF Bay Area, probably has the same market value regardless of where it is. Location is important in appraising - just like many other property attributes such as GLA and Lot Size, elevation, view and so on. But two houses could be essentially in the same location and have much different values due to their other features. To say location is "fundamental" is kind of ridiculous. So are other features.

3. Again what does the Zillow vaiuation model have to do with anything. I certainly never use AVMs for appraising - although they may be used in discussion if better information is not available on some property under discussion. It does seem that you think that someone using regression is using an AVM. That is really far off base.

4. You talk about Mass Appraisal. But what does that have to do with anything? You apparently don't understand what Mass Appraisal is. You confuse it with the use of regression.

I could explain all of these things to you again and again. But, you just don't get it. You are that retrograde.

If I am wrong - give me a correct definition of Mass Appraisal and AVMs. See if you can do that.

But don't feel bad. A certain MAI that should be an expert on resdential appraisal told me a couple of days ago that because I was using MARS, I was doing Mass Appraisal. No! He was being a bit lightheaded.

I never talked about mass appraisal - see my posts.
I agree with you that the MAI believing MARS for an individual appraisal meant you were doing a mass appraisal makes no sense.