When the U. S. Census Bureau began looking into seasonal adjustment of the Quarterly Services Survey series, one question was whether trading day adjustments were reasonable. The series started in the fourth quarter of 2003, and consideration of trading day adjustments began when series had only 17 quarters of data. Beyond concern over series length, however, we questioned more generally whether trading day effects are measurable in quarterly series. To answer our questions, we started with monthly Census Bureau series known to have significant trading day effects. We summed to quarterly levels and tested for significance. We tested using both real and simulated effects. This paper discusses the results of our research.
seasonal adjustment, simulated data, time series
This symbol indicates a link to a non-government web site. Our linking to these sites does not constitute an endorsement of any products, services or the information found on them. Once you link to another site you are subject to the policies of the new site.