|
Activity Number:
|
101
- Time Series Modeling: Mixed Frequency Data, Seasonality, and Model Identification
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 8, 2022 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Business and Economic Statistics Section
|
|
Abstract #323332
|
|
|
Title:
|
A Bayesian Marked Point Process Model for Seasonality in Mixed Frequency Data
|
|
Author(s):
|
Anindya Roy* and Tucker McElroy
|
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau
|
|
Keywords:
|
Seaoanal adjustment;
Measure transformation;
Intensity measure
|
|
Abstract:
|
In the modern data-driven world, data publishing agencies are publishing data at an increased frequency. This has resulted in several published mixed frequency time series data where some parts of the series are at a higher frequency than the other. Seasonal adjustment of mixed frequency time series is a challenging problem. We use a marked point process model with periodic intensity to model activity at the highest frequency level. We use a transformation of the intensity measure to make the model intensity free of periodicity at different levels of aggregation, thereby adjusting the data to be non-seasonal. The method is illustrated with daily, weekly, and monthly time series.
|
Authors who are presenting talks have a * after their name.
