Friday, February 03, 2006

Housing Data Site

I've posted this person's housing data before, but he now has a new site and so I thought I'd bring it to your attention. I will also link to it over in the right margin. It can be found here. The data that includes Marin County can be found here. Enjoy!

And here is a hacked version of his data showing various predictions.


According to this (thanks to the reader who sent it in), panic may have already started in some corners of the world-wide property bubble. What will be, will be...

10 Comments:

Blogger Marinite said...

Yes, my MS Paint skills reign supreme!

Feb 3, 2006, 11:09:00 AM  
Anonymous Anonymous said...

This graph reminds me of a similar graph -- that of the Nasdaq's long term trend up to about 1999....and we all know what happened there......

Feb 3, 2006, 3:10:00 PM  
Anonymous Anonymous said...

If your trend line is defined as constant percentage year-over-year increase, shouldn't it be curving upward, rather than a straight line?

Feb 5, 2006, 5:32:00 AM  
Blogger Rob Dawg said...

Of course the trend line should be compunded. i've been tryin to tell people using these data that same thing for months. They aren't going to change because it reduces the bubble excess by roughly 40% so it doesn't look as scary.

Feb 5, 2006, 8:25:00 AM  
Anonymous Anonymous said...

Perhaps people should learn that an index is a log based graph so a constant percentage growth rate isn't graphed as an upward curve but rather as a straight line...

Feb 5, 2006, 11:15:00 AM  
Blogger poguemahone said...

I'm the "author" of the graphs. Thanks for the comments.

The trend line that I made has been causing a lot of confusion -- and I'll admit, there is probably better way of doing it. In fact, these suggestions are quite useful -- keep them coming.

So, for the time being, I changed the graph a bit and just plotted the year-over-year and quarter-over-quarter appreciation instead of trying to establish a trend.

I also plotted some (arithmetic) moving averages. Maybe a better way to estimate how over-valued or undervalued the hpi is to compare it to its moving average? I know this is done in the stock market quite a bit.

Robert, I'm not sure how to construct a compounded trend line. Do you know of any references? Maybe I need a TA book.

I could take the log of the hpi (y-axes) and THEN do a straight-line fit. Would that be more appropriate?

Feb 5, 2006, 12:20:00 PM  
Blogger Rob Dawg said...

Log Y is the easiest certainly.

When I did the graph for my area Oxnard-Thousand Oaks- Ventura, CA I used simple compounding:

Y1 = Y0 times 1 to power (1 + percent trend)
I then normalized to 1995 = 100.

Moving averages won't give a baseline and will massively lag then overshoot in a bubble.

Feb 5, 2006, 2:37:00 PM  
Blogger poguemahone said...

Robert said:

Y1 = Y0 times 1 to power (1 + percent trend)
I then normalized to 1995 = 100.


Hmm, not sure I understand that. Is this what you are saying?

y(current qtr) = y(previous qtr)^[(1+hpi(current qtr)]

Feb 5, 2006, 4:51:00 PM  
Blogger poguemahone said...

FYI, the y-axis is now log-scale. I agree, this is a better way to look at the data.

Feb 6, 2006, 10:22:00 AM  
Anonymous Anonymous said...

If the trend is constant 6% year-over-year increase:

y(current year)
= y(last year) * 1.06
= y(1995) * 1.06 ^ (current year - 1995)

If you want to do it quarterly, find the quaterly growth rate by taking 1.06^(1/4):

1.06^(1/4) = 1.01467

y(current quarter)
= y(base quarter) * 1.01467 ^ (number of quarters elapsed)

Feb 6, 2006, 10:24:00 AM  

Post a Comment

Links to this post:

Create a Link

<< Home

Terms of Use: The purpose of the Marin Real Estate Bubble weblog (located at URL http://marinrealestatebubble.blogspot.com/ and henceforth referred to as “MREB” or “this site”) is to present and discuss information relating to real estate and the real estate industry in general (locally, state-wide, nationally, and internationally) as it pertains to the thesis that recent real estate related activity is properly characterized as a “speculative mania” or a “bubble”. MREB is a non-profit, community site that depends on community participation and feedback. While MREB administrators do strive to confirm all information presented here and qualify all doubtful items, the information presented at MREB is neither definitive nor should it be construed as professional advice. All information published on MREB is provided “as is” without warranty of any kind and the administrators of this site shall not be liable for any direct or indirect damages arising out of use of this site. This site is moderated by MREB administrators and the MREB administrators reserve the right to edit, remove, or refuse postings that are off-topic, defamatory, libelous, offensive, or otherwise deemed inappropriate by MREB administrators. You should consult a finance professional before making any decisions based on information found on this site.

The contributors to this site may, from time to time, hold short (or long) positions in mentioned and related companies.