Constructing and Modifying Sequence Statistics for relevent Using informR in ��

Area of Science:

• Social Network Analysis
• Computational Social Science
• Statistical Software Development

Background:

• Analyzing complex event histories, particularly sequences of social actions (a→b), has been historically cumbersome.
• Existing methods for relational event modeling often require complex data structures, posing a barrier to researchers.
• The egocentric generalization of Butts' (2008) relational event framework necessitates user-friendly tools for sufficient statistic construction.

Purpose of the Study:

• To introduce the informR package in R, designed to streamline the analysis of complex event histories.
• To provide accessible tools for constructing sufficient statistics required by the relevent package's rem() model.
• To demonstrate the utility of informR with real-world data from the American Time Use Survey.

Main Methods:

• Development of the informR R package, offering user-friendly functions for data preparation.
• Implementation of tools to simplify the creation of complex lists of arrays for rem() model fitting.
• Application of informR to egocentric event data, accommodating multiple event types and support constraints.

Main Results:

• The informR package successfully simplifies the construction of sufficient statistics for relational event models.
• The package handles various complexities, including egocentric event data, multiple event types, and support constraints.
• Illustrative examples using American Time Use Survey data showcase the practical application and ease of use.

Conclusions:

• The informR package significantly lowers the barrier to entry for analyzing complex social action event sequences in R.
• It provides a robust and user-friendly solution for researchers utilizing relational event modeling.
• InformR enhances the accessibility and efficiency of social action analysis in computational social science.