Online Program Home
  My Program

Abstract Details

Activity Number: 394 - Time Series, Reconciliation, and National Statistics
Type: Topic Contributed
Date/Time: Tuesday, August 1, 2017 : 2:00 PM to 3:50 PM
Sponsor: Government Statistics Section
Abstract #324404 View Presentation
Title: Seasonal Adjustment Subject to Accounting Constraints
Author(s): Tucker McElroy*
Companies: U. S. Census Bureau
Keywords: Benchmarking ; Reconciliation ; Indirect Adjustment
Abstract:

For a variety of causes, the indirect seasonal adjustment obtained by aggregating component seasonal adjustments may be inadequate, whereas the direct adjustment of the aggregate can typically be ensured to be adequate by modifying program parameters. Reconciliation techniques can be used to allocate the discrepancies between the direct and indirect adjustments of the aggregate unto the various component series, essentially enforcing that the indirect procedure yields the same outcome as the adequate direct procedure. This paper proposes utilizing adequacy of the component seasonal adjustments -- given the modifications entailed by reconciliation -- as an additional constraint to the accounting problem, using the fact that a broad range of parameters and specifications in a seasonal adjustment method can typically yield an adequate adjustment. The ultimate goal is to reconcile seasonal adjustments arising from the X-12-ARIMA method, evaluating according to the Visual Significance criterion, and apply this constrained reconciliation procedure to the composite series making up U.S. GDP.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association