I am looking for a database that has longitudinal microdata for housing migration patterns.

Optimistically, the data would look something like:

In 2006, Person #2341 made $X, was Y years old, and lived in a Z year old house. In 2007, Person #2341 moved houses.

This way, I can see if we can use microdata to predict if a person is going to move within a year using his information currently via a logit regression.

The ACS database almost works, but it is ex post facto; it tells you the person moved a year ago (so I would be using a persons current information to predict if he moved LAST year; not what I want). I want to know if the person will move the following year, and theres no way to do this using ACS since it doesnt track the same person year to year.

  • $\begingroup$ You could search any paper that has addressed the issue and see what kind of data is used. $\endgroup$ – Héctor Garrido Jul 10 '15 at 6:38

The US Panel Survey of Income Dynamics (PSID) has this data, at least if pre-tax income will do.

We use the 1999-2009 Panel Survey of Income Dynamics to estimate household move probabilities as a function of, among other things, current housing equity. The lock-in effect supposes that mobility decreases with equity, particularly as equity becomes negative. We find that while owners do move less than renters, the move probability increases as homeowners become underwater. The propensity to move out of state in particular increases dramatically for sand state homeowners who have negative equity. There is no lock in effect from negative equity.

Mobility and Mortgages: Evidence from the PSID (Colson and Greece (2013))

The NLSY may also have this data.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.