Standardizing Time in Payroll Data Using Splines

• Activity Number: 134
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Government Statistics Section
• Author(s): Jack Lothian*
• Companies: Statistics New Zealand
• Keywords: Benchmarking ; Calendarization ; Administrative data ; Cubic splines ; Payroll data

Abstract

Statistics New Zealand is in the midst of implementing its 2010-2020 Strategic Plan that will transform how the agency functions. The "administrative data first" philosophy is a critical component in the transformation process and the Business Payroll Tax (PAYE) data is an important input dataset for business and national accounts surveys. Standardized PAYE data can be used in sub-annual surveys to enhance survey data, improve editing and implement calibration. These processes could potentially reduce response burden and collection costs plus improve quality. However, the use of administrative data poses major challenges. The data covers a melange of varying and overlapping time intervals. We propose a calendarization method based on interpolating the cumulated flows with splines that provides data with standardization time intervals and short-term forecasts. The methodology improves the timeliness and quality of the PAYE data and increases the willingness of the survey programs to embrace tax data.