Regression for GLA Adjustment

[NJAPRAZE]

anyone use single / linear regression analysis for determining GLA adjustment (after netting out site and then using cost/sf as the independent variable)?
thoughts on reliability and process?

 

[Jerry Bone Jr]

anyone use single / linear regression analysis for determining GLA adjustment (after netting out site and then using cost/sf as the independent variable)?
thoughts on reliability and process?

The idea is that after all other adjustments are made (when the only difference is GLA) it's a fairly simple modified matched pair analysis.
On the x-axis is GLA, with ASP (adjusted sale price) on the y-axis.

 

[J Grant]

I just extract riht ghere on the grid for all other features and whatever is left is the $ pers sf contrcutoiry value, I might modify it to narrow the value range/ relate it more in sync with a relationship to depreciated cost. It is simpler and more direct that way, and replicates what the market is returning -idk why regression has to be done to get there.

The contributory value of sf is different than the price divided by a GLA sf ( including then the lot value is subtracted out )

 

[alebrewer]

I generally don't isolate the GLA as a single variate analysis. There are too many other variables that are statistically significant (for most data sets) to elicit meaningful results from a single variate analysis - at least in my market.

 

[J Grant]

I generally don't isolate the GLA as a single variate analysis. There are too many other variables that are statistically significant (for most data sets) to elicit meaningful results from a single variate analysis - at least in my market.

If you don't isolate the GLA, then how do you know if it is commanding a premium or not?

Any meaningful difference in GLA sf is not a statistical "variable" , it is a key value driver of a property - and adjusted for in nearly all appraisals I've seen.

 

[alebrewer]

If you don't isolate the GLA, then how do you know if it is commanding a premium or not?

Any meaningful difference in GLA sf is not a statistical "variable" , it is a key value driver of a property - and adjusted for in nearly all appraisals I've seen.

Either you don't understand statistics, you just want to start a fight, or you aren't taking time to read the post.

I never said I didn't account for GLA as ONE OF the independent variables in a multi-variate analysis. What I said was that I generally don't isolate GLA as a SINGLE VARIATE analysis.

Continue on with your blather.

 

[CGinMN]

I adjust for all the other known variables, then do a sensitivity analysis on the GLA.

 

[sputnam]

I rarely use a single method for determining an adjustment. Regression is a useful tool but, it's not the be all and end all of data analysis. There are at least 30 methods for extracting market reaction. The 'best' ones for any particular adjustment depend on the quantity and quality of the data that's available to you.

 

[George Hatch]

I also mostly use sensitivity analysis, sometimes with expanded datasets when necessary. It's easy to explain and easy for the readers to understand, which that's who I'm writing reports for. If I'm lucky enough to score a matched pair in there somewhere (occasionally happens in the prior sales history) then that's additional.

 

[Terrel L. Shields]

riht ghere on the grid for all other features and whatever is left is the $ pers sf contrcutoiry

wipe your glasses JG

The easy way is sensitivity analysis, then copy as picture and paste into the report.