Analysis of Revisions to Monthly Seasonally Adjusted Estimates of New Residential Sales Statistics
Each month the US Census Bureau publishes preliminary estimates of New One-Family Houses Sold. The US Census Bureau releases these estimates to provide government and private data users with early measures of new residential sales activity. A necessary part of the process of issuing these early data involves the issuance of subsequent revisions. The revisions to new one-family houses sold estimates are primarily the result of the replacement of imputed data with data which are reported in subsequent months. New residential sales have larger revisions than other residential construction series. This is due to the fact that most of the new residential construction survey data are based on a permit being issued. Since many homes have a sales contract signed prior to a permit being issued, an estimate must be determined for these sales prior to permit authorization. The "Range" shows the largest decrease (or smallest increase if no months
revised down) and largest increase (or smallest decrease if no months revised
up) over the 12 months. The "Average" is the arithmetic mean of all 12
months. Ideally, this would be zero if the positive and negative revisions
equaled each other. The "Absolute Average" is the arithmetic mean of the
absolute value of the revision. This indicates the average size of revision
without regard to sign.
[PDF] or denotes a file in Adobe’s Portable Document Format. To view the file, you will need the Adobe® Acrobat® Reader available free from Adobe. Analysis of Revisions for New Residential Sales in PDF Format. For questions, visit our Question and Answer Center at ask.census.gov or contact the Residential Construction Branch at (301)763-5160. |