Online Program Home
  My Program

Abstract Details

Activity Number: 168 - Bayesian Models for Gaussian and Point Processes
Type: Contributed
Date/Time: Monday, July 31, 2017 : 10:30 AM to 12:20 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #323590 View Presentation
Title: Handling Missing Data in Self-Exciting Temporal Point Processes
Author(s): James Tucker* and John Lewis and Kathy Simonson and Johnanthan Lane
Companies: Sandia National Laboratories and Sandia National Laboratories and Sandia National Laboratories and Sandia National Laboratories
Keywords: bayesian ; hawkes process ; missing data ; self exciting point process
Abstract:

Self-exciting temporal point processes, such as Hawkes processes, are present in many application areas, including epidemiology, seismology, and surveillance, and require a consistent obser- vation period to accurately model. However, during the observation period, intervals can occur where data is not collected, either through design or malfunction. These intervals of missing data can be problematic, as they can e?ect the modeling of the point process. In this work, we present a Bayesian estimation technique to model univariate, self-exciting temporal point processes with known intervals of missing data. This technique provides a more accurate estimation of the intensity function and provides prediction-informed sensing. We compare our model to current models on simulated data and data from the Global Terrorism Database and show predictive performance improvement.


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

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association