April Seasonal Boost in Home Prices According to the S&P/Case-Shiller Home Price Indices

Data through April 2011, released today by S&P Indices for its S&P/Case-Shiller Home Price Indices, the leading measure of U.S. home prices, show a monthly increase in prices for the 10- and 20-City Composites for the first time in eight months…

The posts on this blog are opinions, not advice.
Please read our disclaimers for Ratings Services, Indices, Equity Research, Securities Evaluations and Risk Solutions.


  1. Mark Morgan says:

    In looking at the April index figures for the Chicago market, each of the tier indices are down more than the aggregate index for the Chicago market. It seems to be the case for both sa and non-sa. How can this be the case? Thank you!

    • Maureen Maitland says:

      Hi Mark,

      In Chicago, low-tiered home prices fell by 3.8% in April over March, mid-tiered by 0.9 and high-tiered prices were up 0.3%. The aggregate index was down 0.4%. So, we are not seeing the results you have stated.

      However, on rare occasions when this does happen it is a technical phenomenon that has to do with the way the repeat-sales model works.

      The tiered home price indices are estimated independently from each other and the aggregate home price indices. This means that there is nothing that constrains changes in the aggregate index to track within the range of changes in the tiered indices. So, although it doesn’t happen very often, it is possible for the month-to-month change in an aggregate index to fall outside the range of changes in the tiered indices.

      There are two main reasons why this might happen: 1) since they are calculated using approximately one-third of the aggregate sample, the individual tiered index points are estimated with less precision and, consequently, they are more volatile from period to period. And 2) the relative weighting of pairs in the tiered and aggregate index calculation is not the same. In particular, when the robust weights are assigned to repeat sale pairs to control for outliers in the data sample, the pairs that the model identifies as an outlier in the aggregate sample may be different than the outliers identified in the estimation of the tiered indices.

Post a Comment

Thank you for submitting a comment. We ask you to use the comment guidelines to promote thoughtful and productive discussions. Your comment will be approved before it will be posted. Thank you for your patience.

Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

  • Categories

  • Recent Comments

  • Tags