187 – Contributed Poster Presentations: Business and Economic Statistics Section
A Mathematical Optimization Approach to Balancing Time Series: Statistics Canada's GSseriesTSBalancing
Michel Ferland
Statistics Canada
Susie Fortier
Statistics Canada
Joana Bérubé
Statistics Canada
Time series data produced by National Statistical Offices and Systems of National Accounts must often respect a vast array of accounting relationships. As the data may come from various sources or undergo non-linear data processing such as seasonal adjustment, the accounting relationships must often be restored before publication. The process used to restore the accounting coherence in the data is referred to as balancing or reconciliation. This paper introduces the GSeriesTSBalancing SAS macro, a numerical optimization solution based on SAS/OR for balancing a system of time series and the latest addition to Statistics Canada's G -Series software, formerly known as Forillon. The mathematical optimization problem is presented and illustrated with simple examples. A comparison with the TSRaking procedure, an alternative regression-based solution available in G -Series for simple balancing problems, is also included.