How To Support An Adjustment Using Multiple Trendlines On A Graph? :

[Inland Insight]

Hello All,
I came across a similar graph while reading The Appraisal of Real Estate, 13th Edition, on pg. 337, and was intrigued with the premise behind it. It seems you can use this graph to support a quantitative adjustment for location. However; I am not sure what the vertical line represents. Does it represent the average or median difference in value, or does it represent the vertical line of best fit between the two data sets, I am not sure?

Secondly, why does the vertical line used land in the center of the graph? I added the vertical line using the insert shape tool in excel, and it I tried to follow the same placement as the original vertical line in the book.

Perhaps someone is familiar with this method of supporting an adjustment and can be of assistance on what it means and how to add a dynamic vertical line that moves as you add or remove data?
Thank you in advance for any input.

 

[Michigan CG]

The line is simply part of the legend and is telling the reader about the gap.

 

[hastalavista]

It looks to me like RR/FRWY warrants an upward adjustment of 10+/-%.
In fact, I do something very similar to this and would narrate it like:

The data supports an adjustment in the range of 8% to 15%. I've selected a 10% adjustment as being reasonable and consistent with the market.​

Done.

(You don't have to put in the vertical line. I'll add stuff if I think it adds to clarity of the message I'm trying to transmit. Otherwise, anyone should be able to look at that chart and see the consistent gap between the $/sf of the two location-types.
And, I know this is an "example" but that relationship is constant; doesn't matter when the sales occur, the difference- a a percentage- is constant.)

 

[Inland Insight]

Denis,

First off, thanks for the excellent response. Secondly, how did you come up with the 8% to 15% range, was it a w.a.g (wild a** guess). Or are you able to look at the graph and pull those percents?
Lastly, this maybe a dumb question (i.e. disclosure), but if the trendlines cross is the difference still constant?

 

[hastalavista]

Secondly, how did you come up with the 8% to 15% range, was it a w.a.g (wild a** guess). Or are you able to look at the graph and pull those percents?

This chart is fairly simple. Pick a point in time and then look where the two lines intercept that point. You can eye-ball the difference.
If you had the equations for each line, you could solve for any point and then mathematically calculate the difference. But I don't think you need to go that far.

Lastly, this maybe a dumb question (i.e. disclosure), but if the trendlines cross is the difference still constant?

No dumb question here, right?

If the trend lines crossed, then the relationship would not be constant but would also change. Right now, those lines look parallel to me, which tells me they are changing at the same rate and the difference at any point in time is going to be the same (constant) on a % basis (since this is a linear trend line). If they are not parallel, then they are not changing at the same rate and the difference is not going to be the same at Time X as it is at Time Y.


IMO, you don't have to be a statistics major to use simple trending to discover patterns in value differences.
This chart is a perfect example. We don't need to solve the regression calculation to see that the value difference is about 8-15% anywhere along the line. And, I would argue that if we did solve the equation and the difference is 11.7%, we'd probably round to 10%.
BTW, the next step to refining the adjustment is to use sensitivity analysis. Again, a not-too-complicated process (CANative constantly reminds me that this is what we do in the adjustment grid when we "tighten up" our adjustments; and he is correct).
So, I said I'd probably reconcile to 10%. But I could reconcile it to 15% if that better fit the data once I start applying adjustments. I could determine that the adjustment might be 5% despite the data in that graph; if I'm going to do that, I should have a good reason which makes sense.

We gather the data, analyze it and see what it implies. I say "implies" because at the end of the day, that is all data does. It implies/suggests/indicates things to us and then we decide how to interpret it. That chart implies an 8-15% location adjustment. I then decide that 10% is the reasonable and market-supported adjustment.

My 2-cents!

 

[hastalavista]

I put on my glasses and zoomed in on the chart; looks more like 15-20%!

 

[Inland Insight]

I agree, 15-20% looks about right. Thanks for your feed back. IMHO, your the man!!!

 

[Terrel L. Shields]

It looks like the lines are not exactly parallel but may be diverging as the SF price changes and that too is logical - the percentage however would remain fairly constant, so at the bottom end of the scale it looks like - 15-20% at 100% and at the top, more like 30% at 200% thus the SF adjustment remains rather constant ratio.