Activity Number:
|
482
|
Type:
|
Topic Contributed
|
Date/Time:
|
Wednesday, August 3, 2016 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Business and Economic Statistics Section
|
Abstract #318815
|
View Presentation
|
Title:
|
Test Based on Frobenius Norm Distance of Spectral Matrices for Presence of Structural Components
|
Author(s):
|
Anindya Roy* and Tucker McElroy
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau
|
Keywords:
|
Structural model ;
Stable parameterization ;
Asymptotic distribution ;
Hypothesis testing
|
Abstract:
|
We propose a test for the presence of a particular component in structural time series models. Specifically, we test for the presence of seasonal components in a structural time series with trend, seasonal and white noise terms. The proposed test is based on the Frobenius norm distance between the model spectral matrix and its empirical value. The test uses a parameterization of the model parameters to ensure stability in the estimated components. The performance of the test is investigated via a limited simulation experiment.
|
Authors who are presenting talks have a * after their name.