tergm
tergm copied to clipboard
statnet
Fit, Simulate and Diagnose Models for Network Evolution Based on Exponential-Family Random Graph Models
Related to but distinct from #33 Currently in the EDA and when targets/monitor is specified by a character, the function `.extract.fd.formulae` is used to attempt to identify the formation, dissolution,...
Added a more user friendly description of package utility, and links to the tergm workshop, 4.0 crosswalk vignette and citation. This updates the README to the new format.
Warnings seem to be related to some deprecated **testthat** functions. Error comes from https://github.com/statnet/tergm/blob/15d8d2dca7250dbc61c6dba2bf3e72ac4511e4f5/tests/testthat/helper-CMLE.R#L213 which should be TRUE but comes out FALSE.
This PR corrects discordBDStratTNT to consider all free edges when checking strat mixing type toggleability. Formerly only non-discordant edges were considered, which in rare cases could lead to incorrect behavior....
Hi Pavel and David, This is Eugene. I spoke to both of you at INSNA conference about my idea of using different weights on past observations and the derived coefficient...
Using `ergm.multi`'s `N()` operator as backend for TERGM CMLE automagically enabled specifying time-varying models. These are now documented, but tests and perhaps examples need to be added.
From https://github.com/statnet/tergm/discussions/120#discussioncomment-9983550 Each TERGM step requires (in principle) an independent draw from an ERGM. The problem is that how many proposals that takes depends on network size, model complexity, and...
Hi @krivit, I made this PoC implementation of the grow-able logging vectors you suggests in #123. Currently it does the following: - make `difftime`, `difftail`, `diffhead` and `diffto` into `DiffVect`....
According to documentation, `tergm(nwd~..., estimate="CMLE")` with `times=` argument omitted should fit one transition, from time 0 to time 1. In actuality, it just crashes: ``` r library(tergm) dummy NetworkDynamic properties:...
`MCMCDyn_wrapper()` pre-allocates three integer vectors (time, tail, head), each of length `control$MCMC.maxchanges`, then trims them as needed. A more memory-efficient approach would be to start with a short vector and...