Mixture Modeling for Temporal Point Processes with Memory

A 15.7 MB research dataset by Xiaotian Zheng, last updated in April 2026, accompanies a paper on constructing temporal point processes with memory. The dataset likely contains synthetic and real data examples used to illustrate a Bayesian inference methodology for modeling event durations with high-order Markov dependence. The work proposes a mixture modeling framework for conditional duration densities to create self-exciting or self-regulating point processes.

Use Cases

• Evaluating Bayesian inference methods for point processes based on the described mixture modeling framework.
• Benchmarking new temporal point process models against the provided synthetic and real data examples.
• Studying high-order Markov dependence in duration processes as described in the modeling approach.
• Analyzing self-exciting or self-regulating behaviors in event sequences using the proposed conditional intensity representation.

Strengths

• Dataset size is 15.7 MB, indicating a manageable download and storage footprint.
• Includes both synthetic and real data examples, as mentioned in the description, for method validation.
• Accompanies a detailed methodological paper proposing a constructive modeling framework.

Limitations

• Row count is unknown, which may limit suitability assessment for large-scale modeling.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality and structure require manual inspection after download.

Provenance

Source

figshare

Collection Method

Likely generated for research purposes, including synthetic data and collected real-world examples.

Freshness

Last updated 2026-04-28 15:13:40; freshness should be verified.