What Variables Predict The Value Of A House?
 

Just because I am a geek, I collected data for 66 possible comparable houses. These were all arguably legitimate comps: same desirable residential neighborhood (all w/in 10 block radius), similar age of house (1910 to 1920s), similar sq ft (avg 3300 sq ft) and #bedrooms (avg 4), similar views (none), all sold w/in past 12 months (4/05 to present).

The data I collected was: date of sale, total listed sq ft, #bedrooms, #bathrooms, sq ft main floors, sq ft finished attic, sq ft finished basement, sq ft unfinished space, sq ft garage, lot size, sale price. Then I ran some regressions to see if there is a "scientific" way to predict the sale price of a house.

I was a bit surprised at which variables were statistically significant and which were not. Care to guess?

Okay, the ONLY variables which were actually statistically significant (t-stat > 2) were date of sale and #bathrooms.

Any real estate pros have a reaction to this? Is this why we see so many new houses with more bathrooms than bedrooms?

P.S. The R-square of the best regression was only 0.55.

Ouija board and offer/demand.

No math can be applied, unless you want to take into consideration imaginay numbers and want to use either Maple or Mathematica to solve some of the equations in a cold Sunday winter afternoon.

After that your guess will be as good as mine...

The only techniques that I know of is typical "gitt'n" price for a comparable home - apprx sq ft, #baths, BRs. If the homes of the same go in less than 30 days then expect a rise of 5% to 7% on the axing price.

Maybe I misunderstand the question. I was under the distinct impression that the three most important factors was - location, location and location..

Next indoor variables likely to be condition of bathroom and kitchen.

Look at the house. Effective age and ammenities are what's missing from the calculations.

recently remodeled homes with the latest upgrades (marble, hardwoods) are gonna be quicker sells.

As well proximity to a good school. Not just any school but a descent school will sell a house faster than many other things. Parents want their kids to go to a descent school and if your house is in that district it really helps.

Definitely a pilot!

rjp

The value of a house is entirely determined by what price someone is willing to pay, simple in theory.

Location and demand.
Demand is the biggie, and is influenced by desire to own and also by expected appreciation, real or not.

GM stock, at ~$20, is trading at 0.78 Price/book, or approximately $25 book value per share. Sometimes the metrics don't pencil out.

Demand has great influence, and we will soon see that supply also has a great influence on pricing.

How did the "date of sales" relate to the prime rate or some other indicator? Higher rates should slow things down a bit.

Curb Appeal!

That's one thing that cannot be measured scientifically.

I've bought and sold a few fixer-uppers, and I always look for something that will be "cute" when done. I want somebody to drive up and decide they need to live in my house.

An ugly house will sell, but only when the buyer thinks they're getting a good deal. That keeps the prices down compared to an attractive house next door.

For most people, buying a house is a very emotional decision. If you want top dollar, you must appeal to the emotions.

Varies by individual. What is a "dump" to some (requiring a lot of sweat equity) and therefore not to their liking, may be just the ticket to someone elase. Some people prefer urban living, some require a more bucolic setting. Some are concerned with travel time. THere are so many factors that once an individual chooses a specific location, for example, giving an opportunity for apples vs apples, then other factors such as condition, size, # of rooms, come int oplay.

Location in JYLs equation is not a factor as it is all within the same hood...

Square Footage as long as it is within the norm for the hood, Bedrooms, Baths and Lot Size...also age which is reflected as condition...and any intrinsic amenity..like a view or pool adds some value.

U boil all this down into $$$ per sq ft....

Even identical houses in the same area can vary significantly. Some factors off the top of my head:

View
Corner lots - negative
Cul-de-sac lots - positive
Usable back yards - more is better
Traffic
Power lines
Neigbors
Proximity to lower valued or commercial properties

All this makes objective valuation just about impossible. I'm working with a buyer who is an engineer and is armed with spreadsheet covering every possible metric. But his wife has final approval and she could care less about his spreadsheets.

Yup.

rjp

Yep, female factor is BIG. That will throw off R^2 big time.

Portland market was rising during the sample period, so date of sale was a predictive variable.

Long-term rates haven't risen much in past year. Here is a neat animation


http://stockcharts.com/charts/YieldCurve.html

Yes! I've always thought corner lots were less desirable. Damn, I should have included variables for corner/non-corner lot and busy/quiet street in my spreadsheet.

THe house I'm currently looking at is on a corner lot, on a semi-busy street, w/ almost no back yard . . . but in a prime neighborhood and great curb appeal.

So if I buy a used outhouse and throw it in the backyard my property value will go up?

Last post then back to work -

One reason I did the study is because the houses here seem to be listed using total sq ft, including finished/unfinished basement sq ft and finished attic sq ft.

My theory was that finished basement sq ft should count for less than main floor sq ft, since it looks like a cave no matter what; that finished attic sq ft should count for less than main floor sq ft, since its odd-shaped and often only usable as a rumpus room for short people; that unfinished basement sq ft should count for much less since its bare concrete and usually lower ceilings; so that main floor sq ft should be more predictive than total (finished and unfinished) sq ft. To my surprise, not so - they were equally insignificant, statistically speaking.

Anyway, I used my regressions to estimate a range of values for this particular house, despite the questionable statistical validity, and find it is listed at $100K over the estimate value range. Plus it is, as I said, a corner lot on a semi-busy street, and also at the periphery of the highly desirable area rather than right smack in the heart of it. But its got that curb appeal.

Check corner lot property taxes. Some municipalities charge by front foot for certain services.

Hmmm. Now my wife has thrown "crime" into the mix.

Her friend from an OR beach town drove her to the neighborhood today ( they were measuring commute times) and said you must'nt live here, the bad area of Portland is too close, you have to be on the west side of the river to be safe.

So I went to the City's website and pulled down the crime stats for the 4,600 person, 430 acre neighborhood. In the past 12 months: 0 homicides, 0 sexual assaults, 2 assaults, 3 street robberies. 11 vehicle thefts. The most common crime is burglary at the rate of 0.8 per 100 persons per year, probably about 2 per 100 houses per year.

I compared this crime rate to our old Berkeley Hills neighborhood, which hasn't had a homicide or sexual assault in over a decade - it is effectively a zero-crime area (unless you are a vehicle or an unoccupied dwelling).

The crime rate per 100 persons is the same in these two neighborhoods.

Indeed, the crime rate in a west-of-river neighborhood we also like is actually higher (though still very low).

Being of analytical mind, I consider the issue resolved. Yet my wife continues to mumble fearfully - "honey, maybe we should live in the suburbs after all" . . .

Oh well. If we end up in some white-bread minivan subdivision, I'm going to make her a Stepford wife.