Ouch! Explaining Home Price Decreases

To the left (in the small print, I apologize) is a table showing median home values for cities in Los Angeles County.

For the most part, what we see here is a decrease in price from 2006 to 2007. This shouldn’t be a surprise to anybody following the news lately. What is of interest is why the drop in price was not consistent. Why did home prices plummet 34% in Pacoima while rising 29% in Santa Monica? Is there something about these regions that explain these price differences, or could it be human error or some other mundane flaw in the data?

Part of the problem we have in breaking these numbers down has to do with the nature of the data, it is aggregate data. (What is listed here in the table is all I have to work with.) This means that instead of the actual data points, instead I have to work with averages. Thus, it would be difficult to do any sort of hedonic model attributing the change in price to characteristics of the house such as gross square footage, number of bathrooms or school quality for instance. While it is possible to back these factors out of the data using some elaborate process (such as an ecological inference solution), I don't recommend it. We will use the data at hand, limitations and all, and see what kinds of valid conclusions can be obtained.

Coming Up Next ...

Chart of the Information and What It Tells Us
Map of the Information and What It Tells Us
Regression and
Forecasting?

Stay Tuned.