Online Program Home
My Program

Abstract Details

Activity Number: 246 - Bayesian Nonparametrics
Type: Contributed
Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #304237 Presentation
Title: Bayesian Dependent Functional Mixture Estimation for Area and Time-Indexed Data
Author(s): Terrance Savitsky*
Companies: Bureau of Labor Statistics
Keywords: Spatio-temporal modeling; Gaussian Markov Random Field; Fourier basis; Bayesian hierarchical models;; Nonparametric statistics; Functional data estimation

The U.S. Bureau of Labor Statistics (BLS) publish employment totals for all U.S. counties on a monthly basis. BLS use the Quarterly Census of Employment and Wages, where responses are received on a 6 ?? 7 month lagged basis and aggregated to county, and apply a time series forecast model to each county and project forward to the current month, which ignores the dependence among counties. Our approach treats these by- county employment time series as a collection of area indexed noisy functions that we co-model. We include predictor, trend and seasonality terms indexed by county. We demonstrate that use of a Fourier basis to model seasonality outperforms a locally- adaptive, intrinsic conditional autoregressive prior on our collection of time series. County-indexed parameters of the 3 terms are drawn from a dependent Dirichlet process (DDP) prior to allow the borrowing of information. We show that use of both spatial and industry concentration predictors produces better prediction accuracy. Our DDP prior accounts for the possibility that nearby counties may express distinct underlying economic structures. Our joint modeling framework computes efficiently to support production.

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

Back to the full JSM 2019 program