Sequence of Percentage Adjustment in the Sales Comparison Approach

[Michigan CG]

.....But if so, why doesn't ANYBODY do it? I mean NOBODY NOWHERE. ......

Just because you haven't seen it doesn't mean people don't do it.

As to narrative explaining what you have done, whether it be straight line or %....you should be explaining what you are doing in a narrative anyway.

 

[Ken B]

Ken B: I think that yours is the ideal approach, indeed. But if so, why doesn't ANYBODY do it? I mean NOBODY NOWHERE. The standard software doesn't offer that option. To calucluate the adjustments manually in the form approach would be feasable but how many page(s) of narrative explanation would be required to explain the approach to a lender when the remaining 99% of appraisals are done with each line item calculated, absolutely, relative to the selling price, rather than sequentially and relative to the revised net adjustment.

I agree with you in principle that the method you proffer is THE only way to make adjustments appropriately, although I am unable to explain why it ISN'T DONE by ANYBODY EVER.

I guess you missed the part of the post where I identified that I just used percentage adjustments in a comp grid this afternoon. Of course, it was a grid I created on an Excel worksheet as part of a narrative appraisal of another dead town home subdivision I am appraising. Just one of many I have done and will do (comp grids using percentage adjustments in Excel and appraisals of dead town home subdivisions, that is).

Now, the only adjustment I did not make in the comp grid was for differences in GLA. This is due to the fact that I analyzed the comps on a price per square foot basis. After adjustment, the adjusted sale price per square foot for the comps ranged from around $110 to $150. Analysis of the data revealed an inverse relationship between GLA and price per square foot. This was expected given economies of scale which would normally result in decreasing price with increasing size. (Think of the cost of a single bottle of soda purchased from a convenience store and the cost per bottle of soda when purchasing a case of sodas at Costco. The difference in price is a result of economies of scale.)

Creating a scattergraph using the GLA as the independant variable and the price per square foot as the dependant variable, there was an obvious trendline which, when graphed, indicated a value per square foot for the typical unit in my project of $127. The correlation coefficient was .952 indicating high correlation in the data. I could have removed an outlier to increase the correlation coefficient, but decided it was unnecessary. As it was, I used 8 comparable sales in the grid. That was the least number I felt was adequate to provide a reasonable indication of value.

Bet you never use any of that in your reports either, huh?

 

[ZZGAMAZZ]

1) Granted the inverse relationship between GLA and $/sqft although I can't remember the verbiage that that Santori (sp?) used last summer to describe the distinction between price-per-unit and price-per-sqft-per unit...

2) "Confounding" factors as they're typically called in the parlance of statistical analysis might be the basis of the strong positive correlation, kinda like studying the points per game of white guys versus black guys but forgetting to mention that the black guys are taller and more athletic, and rebound better and can dribble the ball as well . . . that is to say, if the 2 variables proved to be inversely related why would you bother to flip them unless you had a preconceived notion, which goes to faulty research "design" and precludes the significance (in non-statistical terms that is) of the outlyers being included or excluded (although I must admit that the wherewithal of the researcher to exclude outlyers was an epiphany of sorts to my small dull brain, at which point in time I finally--at a late age--realized scientifically why ugly girls don't get asked to dance...)*

3) Sincerey, I agree with your notion of the importance of relative adjustments made in the appropriate declining sequence; however, a) I don't think ANYBODY ANYWHERE does it routinely (and I know I'm right because I get no props from the Forum and I'dda been blasted 24 hours ago if I was wrong; and b) more than anything the arguement goes to the concept of "Intended User" because, respectfully speaking, I'm equally as positive that you're playing with yourself if you think ANYBODY ANYWHERE cares about the erudite analysis, which isn't being presented as a case study at an appraisal convention and has little practical added value...

*does a set of parenthesis inside another set of parenthesis, in a string of text, require the outside parentheses to be brackets rather than parentheses, or am I having a flashback to my 1st (of way too many) stats courses.

--GM, AR. AA. BS. MSJ. MBA. MHA (latter pending...and many more to follow)

P.S. Sorry I took so long to respond but I have a day job that requires approx 10 hours/day and I barely had time to finish the inspection of a 9000 sqft SFR...before dark...

 

[Ken B]

LOL...too funny.

MY POINT was...just because you haven't seen something done in a form, that does not mean that NOBODY ANYWHERE is doing it. I demonstrated analysis that you likely do not see in forms as further evidence that there is a big world outside of Formland.

I would suggest if you would like to test your hypothesis that NOBODY ANYWHERE uses percentgage adjustments, you should take it to the Commercial sub-forum.