New Gaussian Process model enhances analysis of self-exciting count data

Researchers have developed a new statistical model called the Gaussian Process Discrete Hawkes Process (GP-DHP) designed for analyzing discrete-time count data where past events influence future occurrences. This semiparametric model utilizes Gaussian-process priors for both baseline and excitation components, allowing for the estimation of diverse excitation shapes and evolving baselines. The GP-DHP has demonstrated superior or comparable predictive accuracy across various real-world datasets, including disease surveillance, shooting incidents, and terrorism events. AI

IMPACT This model offers improved analytical capabilities for time-series data with self-exciting properties, potentially benefiting fields like epidemiology and public safety.

RANK_REASON The cluster contains a research paper detailing a new statistical model. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv stat.ML →

• Gaussian process
• Gaussian Process Discrete Hawkes Process
• Gun Violence Archive
• Hawkes Processes
• New York City
• Singapore
• Trinnhallen Brisley

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →