ESRI Research Seminar: "Improving the Seasonal Adjustment of the Retail Sales Index using X-12-ARIMA"

Venue: ESRI, Whitaker Square, Sir John Rogerson's Quay, Dublin 2

Speaker: Patrick Foley, Central Statistics Office.

Sub-annual seasonally adjusted macroeconomic statistics provide critical tools for economic policy makers, business cycle analysts and forecasters. Amongst the various economic metrics available, good quality seasonally adjusted monthly retail trade indices provide critical information on the economic well-being of any developed economy. Inappropriate or low-quality seasonal adjustment can generate misleading results and increase the possibility of providing false signals on the economy. The seasonal adjustment of the Irish retail sales figures is, however, complicated by the standardised 4-4-5 reporting period (SRP) used to collect the raw data. This reporting system results in complex atypical calendar effects in the data, making seasonal adjustment of the unadjusted retail sales figures particularly challenging. The X-12-ARIMA program, developed by the U.S. Census Bureau, is used to seasonally adjust the retail sales figures. This paper describes the improved methodology and procedures adopted by the Central Statistics Office to seasonally adjust the retail sales indices.