I suspect no one wants to take the other side to JD's points on regression and its application.
It's all about the money; get appraisers on board but charge them a fee.
Look, by no means did I say that regression could not be used in many cases as an alternative valuation product.
My point was that (A) you can do it yourself since no one owns regression and there are a multiude of free and commercially available tools to do it and (B) you can learn more about it for free within an hour on the Internet than you can by paying someone who can't do it themselves by hand!
Regression is best left to do itself on its own without "tampering" with the data or the variables you want to use. It is BETTER to have a grossly negative number for something like Baths while another variable is grossly INFLATED, which self-levels the analysis, than to be taught to play around with the data and the variables until it "makes sense," which is what one idiot who actually did co-author a book on regression believes should be done. "Results don't look right, take out the fireplaces variable. There, now that looks rght." No, it was right the first time! Such an approach is SKEWING and makes the analysis irrelevant and no better than pulling numbers ot of thin air.
The only thing you should be taught about regression is that the data should be of a similar market (in the case of regression for appraisal), and the available data may not lend itself to provide a well-formed calculation, in which case such an approach is not appropriate. That's all you need to know about it. Period. And both of these items are obvious to even a child.
Regression is an accepted science it is acceptable to use whether you can do it yourself without a calculator or not. And just knowing what it is doing doesn't make the calculation better or worse.
Don't use data that is irrelevant to the subject's market entirely. This does NOT mean do not use homes with varying bedroom, bath, GLA, or site area. In order for a regression analysis to work at all, there MUST be variance in the data. For example, how can any mathmatical forumale tell you what the effect of a bedroom is if all your sample data have the same number of bedrooms? It CAN'T! You WANT variance in your data, but you MUST exclude the "Qualitative" items that are far beyond inclusive.
For example, in my own neighborhood, a 2400 SF house with 4-bedrooms and 2.75 baths in good remodelled condition and on an 8k to 10k SF lot routinely sell between $250k and $350k. However, there is a subdivison within a quarter mile that has slightly newer homes on 5k to 7k lots, 3 bedroom, 2 baths that sell routinely between $450k and $600k Don't ask me why, but because of "perception" of better design (their floor plans are horrible compared with the older homes) and newer quality, they are a separate and distnict market and including these homes would indicate that a bath, GLA, and SITE AREA is worth NEGATIVE, even despite the age difference that can be calculated, because the higher sales have a decreased quantity of all. Does this make sense??? No, and for that reason, you would exclude such sales, but you also wouldn't include them even if you were desperate in a traditional appraisal (at least I hope you wouldn't).
There are so many "qualitative" items that simply can't be taken into account, and trying to rate these on a scale of 1-to-6 like one program does or even a scale of 1-to-100 would not change very much the validity and, in fact, now places personal judgment into a scientific analysis....not good.
The idea is good, but others have had this, you can do it yourself, and there's not much you can be taught about either appraisal OR regression from peope who don't understand either that would make it better or worse. Follow the two steps (appropriate data, quality of data in) and no matter WHAT skill level or understanding of regression the results should be the same.
Seems to me they are just hoping you will promote the idea to your clients so that they get a huge chunk from every assignment you do. What's the benefit to YOU? Yes, maybe get the ball rolling (towards them) so that banks and AMCs start using this instead of a BPO. What then, the AMC isn't going to charge more for this than a BPO, maybe slightly, but then they take their cut and Bradford takes theirs and what are you left with? Who benefitted the MOST? Say they CAN charge $150 (AMCs are paying less than that to appraisers for full 1004s), that means they MIGHT give the appraiser $75, and then what is the cut for using the CVR and the data you could have gotten for FREE? What are YOU left with but FAR less than 50%. And you know $150 per this report isn't going to happen with any kind of frequency. Most likely it will be less than $75 as are BPOs. So, after the AMC and Bradford take their cut, what are you left with?
Or, is Bradford backdooring into an AMC and maye there will be only one split of the BPO pie. Sure, that would leave a little more for appraisers. Oh, that will go down nicely in our industry. As soon as they do, I wonder how long it will take ACI and a la mode to say "we are no longer going to bother to translate their reports so if you want to do business with the thousands of clients that use our service you're going to have to find an alternative." Would they then cry "foul" for the other companies doing exactly what they are attempting to do? Hmmm.
But that doesn't mean you cant use the general concept, which is NOT new or exclusive to Bradford, to compete with BPOs or simply to cover your butt when providing "comp-chekcks." Such a concept is IDEAL for this and this is why you SOULD be employing this in your day-to-day practice. Just don't pay someone for it unless their cut is much smaller than yours and THEY got you the job without YOU having to promote it for them.
That's my point. The topic was about training appraisers. Why pay for the training you can do yourself. Look at the ulterior motives. Train you on a "new" (been around awhile, guys) product, get you all excited about it and send you out looking for work? Could it be to build an army of appraisers promoting THEIR interests rather than the appraisers? Sure, some of you may win a little, but they win a lot due to YOUR promotion. At least the current AMCs do their own promotion to the clients while screwing the appraiser. Here they want the appriaser to screw themselves and PAYto do it in my opinon.
There are other organizations looking out for YOU that are doing the same thing, but THEY are doing the promotion themselves and not raising revenue by "training" you.
Heck, I can put a synopsis of regression, general analytics, interpretation of resls, and even a test on my website that will teach you more than you need to know. Nobody OWNS regression, and certainly no one OWNS the application of regression in appraisal. Only a fool would even think this is a new concept that has never been deployed or utilized by an appraiser.
Good idea. Bad implementation. Waste of money. Another hand in appraiser's pockets!
That's all I'm saying.
community for the discussion of real estate appraisal. Register a free account to be able to post and unlock additional forums and features.
