Online Program Home
My Program

Abstract Details

Activity Number: 565 - Time Series in Government and National Statistics
Type: Topic Contributed
Date/Time: Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
Sponsor: Government Statistics Section
Abstract #307096
Title: Assessing Residual Seasonality in the U.S. National Income and Product Accounts (NIPA) Aggregates
Author(s): Baoline Chen* and Tucker McElroy and Osbert Pang
Companies: Bureau of Economic Analysis and US Census Bureau and U.S. Census Bureau
Keywords: Time Series Analysis; Diagnostic Tools for Detecting Residucal Seasonality; Residual Seasonality in GDP Growth

There has been an ongoing debate in the public sphere on whether residual seasonality is present in the estimates of GDP and its major components published by the Bureau of Economic Analysis (BEA). This topic has stemmed from the observation that in recent years GDP and some of its major components consistently grow at a much lower rate in the first quarter than in the other quarters of the year. These critiques have prompted renewed interest in seasonality diagnostics and seasonal adjustment at BEA. Some changes have been made to address the weak quarter-one growth, but some studies point out continuing difficulties. The findings in the previous studies seem sensitive to the methods and sample span used. This article aims to bring clarity to the topic by summarizing methodologies used; arguing for sound methodological frameworks for evaluating claims of residual seasonality, because if a statistical methodology is applied wherein the chief axioms are violated, any resulting claims should be treated as dubious; and applying our methodologies to different vintages, sample spans, and components of GDP, making comparisons with the results obtained from some of the previous studies.

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

Back to the full JSM 2019 program