X-12-ARIMA

Investigating Quarterly Trading Day Effects

Kathleen M. McDonald-Johnson⁽¹⁾, David F. Findley⁽²⁾, and Erica Cepietz⁽³⁾

ABSTRACT:

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.

KEYWORDS:

seasonal adjustment, simulated data, time series

• Back to Seasonal Adjustment Papers page
• Download PDF file (about 526K)

⁽¹⁾ Kathleen M. McDonald-Johnson is Mathematical Statistican, Office of Statistical Methods and Research for Economic Programs, U. S. Census Bureau, 4600 Silver Hill Road, Washington, DC 20233. email : kathleen.m.mcdonald.johnson@census.gov

⁽²⁾ David F. Findley was the Senior Mathematical Statistican for Time Series and is now a consultant, U. S. Census Bureau, 4600 Silver Hill Road, Washington, DC 20233. email : david.f.findley@census.gov

⁽³⁾ Erica Cepietz is Mathematical Statistican, Decennial Statistical Studies Division, U. S. Census Bureau, 4600 Silver Hill Road, Washington, DC 20233. email : erica.cepietz@census.gov