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.
- Status
Austin
Elite Member
- Joined
- Jan 16, 2002
- Professional Status
- Certified General Appraiser
- State
- Virginia
Terrell: You mentioned S-curves, probably being sarcastic, but I actually discovered S-curves doing regression analysis. I think the way I found them was in appraising cape COD homes. I was doing iterations on cape COD homes and was getting data that formed an S-curve. I couldn’t figure out why. I then did a spectrum analysis of cape COD homes and found out why. Cape COD homes are clustered across the price spectrum. There are clusters of cape cods in this market at around $55,000, a cluster around $100,000, then a cluster around $150,000 then $225,000, and no cape cods in the gaps. The S-curves were caused because there was some critical value factor that was the critical trigger point. Go below a certain point and the regression said it was in one cluster, make a small adjustment and it fell into another cluster. That is why I did the spectrum analysis and discovered the cluster theory and what the critical factor is. Another example of the virtues of regression and the danger of AVM. I have never run into an appraisal problem I could not solve with regression methods. I think all of the criticism of the subject is based on fear of the unknown. The biggest thing you have to fear from regression is that it reveals the shortcomings of the present appraisal method. As far as the statement that math works in Austin’s market but not in yours, that is like Bill Clinton's line: “Trickle down economics doesn’t work.” That is like saying blood doesn’t circulate. In certain areas of the country Clinton might be right, blood doesn’t circulate. These areas are easy to spot; they are colored blue on the election map. Get your hands on a good math book and see what it says about people that extrapolate, then look at the election map and see where they live. Remember my 172,000,000,000-pound grandson story from an above post. If the trend continues, that is what he will weigh in 21 years by extrapolation. According to Terrell, this is a valid projection under USPAP. Anybody want to get up a pool?
Tom Hildebrandt
Member
- Joined
- Jan 16, 2002
- Professional Status
- General Public
- State
- North Carolina
Austin
You know I like regression models and that I think they are an excellent tool, and can provide great insight into a market.
But I think you have carried the point to an extreme. USPAP does not say extrapolation or any other methodology, technique or tool is the right way. USPAP Std 1-1 (a) (lines 512-513) we must "those recognized methods and techniques that are necessary to produce credible results."
A person who extrapolates in the manner you suggest is not using the technique correctly and would not get credible results. A person who uses a statistical model that is not correctly designed for solving a particular problem also will not get credible results.
Some appraisal problems do not have empirical data which can be used to measure the specific value element, they are in fact hypothetical in the truest sense of the word. Empirical models, such as RA, can not provide definitive solutions to such problems, they can possibly provide insight into relevant issues. From those insights we can make judgements.
Judgement in valuation problems is what sets the appraiser apart from a statistician. If mere statistical analysis could solve all valuation problems, there would be no appraisal profession.
You know I agree that statitics, particular multiple linear regression, is an excellent tool and is overlooked by most appraisers, but it is not the end all solution.
Best regards
Tom Hildebrandt GAA
You know I like regression models and that I think they are an excellent tool, and can provide great insight into a market.
But I think you have carried the point to an extreme. USPAP does not say extrapolation or any other methodology, technique or tool is the right way. USPAP Std 1-1 (a) (lines 512-513) we must "those recognized methods and techniques that are necessary to produce credible results."
A person who extrapolates in the manner you suggest is not using the technique correctly and would not get credible results. A person who uses a statistical model that is not correctly designed for solving a particular problem also will not get credible results.
Some appraisal problems do not have empirical data which can be used to measure the specific value element, they are in fact hypothetical in the truest sense of the word. Empirical models, such as RA, can not provide definitive solutions to such problems, they can possibly provide insight into relevant issues. From those insights we can make judgements.
Judgement in valuation problems is what sets the appraiser apart from a statistician. If mere statistical analysis could solve all valuation problems, there would be no appraisal profession.
You know I agree that statitics, particular multiple linear regression, is an excellent tool and is overlooked by most appraisers, but it is not the end all solution.
Best regards
Tom Hildebrandt GAA
Austin
Elite Member
- Joined
- Jan 16, 2002
- Professional Status
- Certified General Appraiser
- State
- Virginia
Tom: What is your definition of judgment? What is your definition of sanity? Good judgment and sanity are best defined as decisions based on the highest probability of success. Insanity and bad judgment are defined as making decisions with a low degree of probability of success and or a high degree of failure. Judgment rests on a foundation of probability theory. Can you give me an example of judgment that does not involve an element of probability? I am not taking statistics to extremes; I am just acknowledging it where it exists. The cost manuals are statistical models; the selection of comparable sales is a statistical model(do you select comparables based on their probability of representing the subject or based on their probability of being way out of the ordinary?); the income approach is a statistical model for the same reason. Do you select income and expenses from the middle of the range or from the extremes of the range? You can’t get away form it, statistical methodology is a major element in human logic. The two major elements of human logic are: How probable is it, and what is it like in comparison. Can you give me an example of using good appraisal judge that does not hinge on some probability of success and or failure? The problem in my mind is not my taking static’s to the extreme, but others denying static’s to the extreme. How do you decide what is comparable and what is not comparable for instance?
Tom Hildebrandt
Member
- Joined
- Jan 16, 2002
- Professional Status
- General Public
- State
- North Carolina
Austin
And your point is?
Judgement is required in identifying the problem to be solved. It is the mental ability to come to a judgement about things, the power of comparing and deciding, understanding and good sense; at least that is Webster's definition and is what USPAP is about. Once we have applied judgement in identifying the problem to be solved, we apply judgment in the execution of developing our opinions relative to the identified problem. That is where competency and experience come in, in the execution of forming the opinions and conclusions necessary to solve the problem.
Sure, all of life is full of judgements as to the probabilty of future events. But that does not reduce all decision making to mere statistical probability of certain events occurring. To argue so belies the variableness of humans in applying judgement consistently and uniformly. You have so argued yourself.
All I am suggesting is that in the reolsution of the more complex appraisal assignments, particularly when fewer sets of empirical data are availble, the more judgement comes into play. In the case of RA, it is in how you identify the population of data, or how you characterize the variables, it is how you interpret the results.
Judgement is what we have to sell regardless of how it you choose to package it.
Best Regards
Tom Hildebrandt GAA
And your point is?
Judgement is required in identifying the problem to be solved. It is the mental ability to come to a judgement about things, the power of comparing and deciding, understanding and good sense; at least that is Webster's definition and is what USPAP is about. Once we have applied judgement in identifying the problem to be solved, we apply judgment in the execution of developing our opinions relative to the identified problem. That is where competency and experience come in, in the execution of forming the opinions and conclusions necessary to solve the problem.
Sure, all of life is full of judgements as to the probabilty of future events. But that does not reduce all decision making to mere statistical probability of certain events occurring. To argue so belies the variableness of humans in applying judgement consistently and uniformly. You have so argued yourself.
All I am suggesting is that in the reolsution of the more complex appraisal assignments, particularly when fewer sets of empirical data are availble, the more judgement comes into play. In the case of RA, it is in how you identify the population of data, or how you characterize the variables, it is how you interpret the results.
Judgement is what we have to sell regardless of how it you choose to package it.
Best Regards
Tom Hildebrandt GAA
Austin
Elite Member
- Joined
- Jan 16, 2002
- Professional Status
- Certified General Appraiser
- State
- Virginia
Tom: That is a good dodge, but is a circular reference. I am not questioning the necessity of good judgment, the question was: “Give us an example from your appraisal experience where you used judgment that was not directly or indirectly related to probability of some sort.” You gave a good answer, but I can’t figure out what the question was. What I am trying to get across, apparently not very well, is that the best tool for exercising good appraisal judgment is statistics. What you seem to be saying is that if you exercise good judgment, then you don’t need statistics. That is like saying: “If I have strong hands, why do I need tools.” I guess you can muscle your way through life if you have strong hands, but why bother when automatic power tools are available to take the pain out of the chore? Then too, we don't all have strong hands.
Tom Hildebrandt
Member
- Joined
- Jan 16, 2002
- Professional Status
- General Public
- State
- North Carolina
Austin
I said in an earlier post that RA is probably the best tool but not the only tool.
My logic is not circular. I maintain that judgement is the cornerstone in the apprasial process and is the first essential elemnt. Without judgemnt you can not adequately define the problem and can not determine the relevant issues. If you have not defined the problem, you can not select the correct tools.
to use your analogy, " I have hands, and I have tools. Tell me what I am building and why, I will use judgement to select the right tools, whether they be power tools or hand tools". If I am building a 100 chairs for sale in the market place and my tarket market is low end buyers, I will use an assembly line and power tools. If I am building a one off high end custom product because I consider myself an artist and craftsman, I may choose antique hand tools."
Since an empirical model can not answer questions for which there are no data within the set, and since much of the what is involved in appraising is subjective (market appeal is a good example), the entire process of developing a model to reflect these subjective issues requires judgment.
This is not circular logic, judgement comes first in the process of appraising. Whether it is good or not depends on the individual appraiser.
Regards
Tom Hildebrandt GAA
I said in an earlier post that RA is probably the best tool but not the only tool.
My logic is not circular. I maintain that judgement is the cornerstone in the apprasial process and is the first essential elemnt. Without judgemnt you can not adequately define the problem and can not determine the relevant issues. If you have not defined the problem, you can not select the correct tools.
to use your analogy, " I have hands, and I have tools. Tell me what I am building and why, I will use judgement to select the right tools, whether they be power tools or hand tools". If I am building a 100 chairs for sale in the market place and my tarket market is low end buyers, I will use an assembly line and power tools. If I am building a one off high end custom product because I consider myself an artist and craftsman, I may choose antique hand tools."
Since an empirical model can not answer questions for which there are no data within the set, and since much of the what is involved in appraising is subjective (market appeal is a good example), the entire process of developing a model to reflect these subjective issues requires judgment.
This is not circular logic, judgement comes first in the process of appraising. Whether it is good or not depends on the individual appraiser.
Regards
Tom Hildebrandt GAA
Austin
Elite Member
- Joined
- Jan 16, 2002
- Professional Status
- Certified General Appraiser
- State
- Virginia
Tom: The economic measure of success is profit. Your example of using hand tools instead of power tools can or cannot be profit motivated. If you are motivated by profit because there is more profit in hand made furniture than in machine milled furniture, then your judgment is based on the probability of more profit based on some form of statistical empirical evidence. If you are using hand tools because you love abstract art forms, that is not profit motivated and thus an abstraction. A rose is abstract art but its value in dollars is subjective. If there is more profit in machine milled furniture and you opt for less profitable hand crafted furniture, then in an economic sense, you exercised bad judgment by choosing that method. Again, bad judgment is defined as making a choice with a lower probability of success or a higher probability of failure.
Then you say, “no data set exist to measure abstractions like market appeal, thus you have to rely on judgment.” That is the circular reference. Given the above definition of good/bad judgment, then what is your criterion for measuring the abstraction of market appeal? Appraisal is an economic enterprise based on profit. To be consistent your judgment must be based on some estimate of profitability of market appeal. How do you decide which market appeal is profitable and which is not? I submit, the judgment is based on statistical data that indicates one appeal is worth zzz$’s more than another market appeal. What I am trying to get across is that the appraisal business will never be accepted as a true profession as long as we hold onto that fuzz factor we call “gut feelings” and “abstract undefined judgment.”
Another example: When I was selling real estate there was always some nut that would look at a hundred houses. I asked a woman once if she had seen anything she liked. She said “no, but I will know what I like when I see it.” I asked her to draw me a picture of the ideal dwelling she was looking for. I never saw her again. Why? Because you can’t quantify abstractions, and if you can’t quantify abstractions, how can anybody know when you find what are looking for and measure its worth? How can you make comparisons when you don’t know what your are comparing? How can judgment get you out of this dilemma?
Then you say, “no data set exist to measure abstractions like market appeal, thus you have to rely on judgment.” That is the circular reference. Given the above definition of good/bad judgment, then what is your criterion for measuring the abstraction of market appeal? Appraisal is an economic enterprise based on profit. To be consistent your judgment must be based on some estimate of profitability of market appeal. How do you decide which market appeal is profitable and which is not? I submit, the judgment is based on statistical data that indicates one appeal is worth zzz$’s more than another market appeal. What I am trying to get across is that the appraisal business will never be accepted as a true profession as long as we hold onto that fuzz factor we call “gut feelings” and “abstract undefined judgment.”
Another example: When I was selling real estate there was always some nut that would look at a hundred houses. I asked a woman once if she had seen anything she liked. She said “no, but I will know what I like when I see it.” I asked her to draw me a picture of the ideal dwelling she was looking for. I never saw her again. Why? Because you can’t quantify abstractions, and if you can’t quantify abstractions, how can anybody know when you find what are looking for and measure its worth? How can you make comparisons when you don’t know what your are comparing? How can judgment get you out of this dilemma?
I enjoy your wisdom, some folks just don't get it 8)Tom Hildebrandt
Member
- Joined
- Jan 16, 2002
- Professional Status
- General Public
- State
- North Carolina
Austin
If you change the definition in an argument, you can make anything sound illogical.
My definition of judgement is from Webster and comes without the economic concept of value, much less that value is profit oriented. The appraisal process requires the appraiser to apply judgement in determining the relation ship between the use and purpose of the appraisal, as well as in establishing the key elements in the definition of value you are seeking.
If in the analogy we are debating, just like in the apprasial process, the individual who intends to make an informed opinon has the obligation to make a judgement of what the necessary components upon which an appropriate decision would be based. If profit is the sole criteria, and hand crafted artifacts do not bring the maximum profit, then there would be bad judgement in selecting hand tools. But if the problem involved consideration of the artistic element, or if there were a question of which would bring the highest profit, then the problem and its solution might well dictate a different conclusion.
I agree with your philisophical argument that all life is a series of choices, and to a degree, the choice is in some manner measuring the good verses bad of the potential outcomes, and to that extent and in that context, all life is statistically oriented.
I do not accept the premise that mere statistics, based on empirical data, provide the best solution.
I would agree that good judgement, coupled with good sound RA techniques, yield superior results to either approach used without the other.
Right now I have to make the judgement that I had better finish the appraisal report now due. The decision is based on two factors, there is a high probability that my client will not pay me if I do not get it to him on time, and secondly, there is an even higher chance that my wife will Kick my b... if I do not get paid ! I suppose that is covariance, but...
Regards
Tom Hildebrandt GAA
If you change the definition in an argument, you can make anything sound illogical.
My definition of judgement is from Webster and comes without the economic concept of value, much less that value is profit oriented. The appraisal process requires the appraiser to apply judgement in determining the relation ship between the use and purpose of the appraisal, as well as in establishing the key elements in the definition of value you are seeking.
If in the analogy we are debating, just like in the apprasial process, the individual who intends to make an informed opinon has the obligation to make a judgement of what the necessary components upon which an appropriate decision would be based. If profit is the sole criteria, and hand crafted artifacts do not bring the maximum profit, then there would be bad judgement in selecting hand tools. But if the problem involved consideration of the artistic element, or if there were a question of which would bring the highest profit, then the problem and its solution might well dictate a different conclusion.
I agree with your philisophical argument that all life is a series of choices, and to a degree, the choice is in some manner measuring the good verses bad of the potential outcomes, and to that extent and in that context, all life is statistically oriented.
I do not accept the premise that mere statistics, based on empirical data, provide the best solution.
I would agree that good judgement, coupled with good sound RA techniques, yield superior results to either approach used without the other.
Right now I have to make the judgement that I had better finish the appraisal report now due. The decision is based on two factors, there is a high probability that my client will not pay me if I do not get it to him on time, and secondly, there is an even higher chance that my wife will Kick my b... if I do not get paid ! I suppose that is covariance, but...
Regards
Tom Hildebrandt GAA
- Status