Missing Statistics in eventnet Output

[Yu Xu]

Dear all,

I am writing to seek your advice regarding an issue I encountered while using eventnet for a relational hyperevent modeling (RHEM) analysis of a collaboration network.
After running eventnet with a fully specified configuration (including node-level attributes, hyperedge attributes, dyadic attributes, and statistics such as repetition, triadic closure variants, and sub-repetition statistics), the output CSV file appears to contain only the base columns (IS_OBSERVED, SOURCE, TARGET, TIME, TYPE, WEIGHT, EVENT_INTERVAL_ID, EVENT, TIME_POINT, TIME_UNIT, EVENT_INTERVAL) with no statistic columns present. Thus, when I tried to fit RHEM in R, I failed...
Could you advise whether there might be another reason why the statistics are absent from the output? Any guidance would be greatly appreciated. I attached my  configuration file below.

Best regards,
Yu

[juergen lerner]

Dear Yu,

This has a simple explanation. The observations in your configuration file are of type "DEFAULT_DYADIC_OBS" (i.e., for dyadic events having one source and one target) but all the statistics are for directed hyperedges ("DHE_..."). Any observation computes and outputs only statistics that are compatible with its type (dyadic events vs undirected hyperedges vs directed hyperedges). Also see:

https://github.com/juergenlerner/eventnet/wiki/Troubleshooting-(help)#confusion-between-dyadic-events-undirected-hyperevents-and-directed-hyperevents

for a discussion of this issue. 

To fix the problem you would have to change the type of the observations. COND_SIZE_DHE_OBS seems to make the most sense.

Best wishes

  Juergen

[Yu Xu]

Hi Juergen,

Thank you so much for correcting that error here! I am very grateful! Wish you a great upcoming week.

Best,

Yu