CRAN Package Check Results for Package ergm.count

Last updated on 2022-01-22 23:52:08 CET.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 4.0.2 10.26 1289.35 1299.61 WARN
r-devel-linux-x86_64-debian-gcc 4.0.2 7.39 221.96 229.35 OK
r-devel-linux-x86_64-fedora-clang 4.0.2 1018.63 OK
r-devel-linux-x86_64-fedora-gcc 4.0.2 478.09 OK
r-devel-windows-x86_64-new-UL 4.0.2 40.00 345.00 385.00 OK
r-devel-windows-x86_64-new-TK 4.0.2 OK
r-patched-linux-x86_64 4.0.2 14.92 261.82 276.74 OK
r-release-linux-x86_64 4.0.2 9.44 287.30 296.74 OK
r-release-macos-arm64 4.0.2 OK
r-release-macos-x86_64 4.0.2 OK
r-release-windows-ix86+x86_64 4.0.2 29.00 448.00 477.00 OK
r-oldrel-macos-x86_64 4.0.2 OK
r-oldrel-windows-ix86+x86_64 4.0.2 28.00 407.00 435.00 OK

Check Details

Version: 4.0.2
Check: re-building of vignette outputs
Result: WARN
    Error(s) in re-building vignettes:
     ...
    --- re-building 'valued.Rmd' using rmarkdown
    Loading required package: ergm
    Loading required package: network
    
    'network' 1.17.1 (2021-06-12), part of the Statnet Project
    * 'news(package="network")' for changes since last version
    * 'citation("network")' for citation information
    * 'https://statnet.org' for help, support, and other information
    
    
    'ergm' 4.1.2 (2021-07-26), part of the Statnet Project
    * 'news(package="ergm")' for changes since last version
    * 'citation("ergm")' for citation information
    * 'https://statnet.org' for help, support, and other information
    
    'ergm' 4 is a major update that introduces some backwards-incompatible
    changes. Please type 'news(package="ergm")' for a list of major
    changes.
    
    
    'ergm.count' 4.0.2 (2021-06-18), part of the Statnet Project
    * 'news(package="ergm.count")' for changes since last version
    * 'citation("ergm.count")' for citation information
    * 'https://statnet.org' for help, support, and other information
    
    Best valid proposal 'DiscUnif' cannot take into account hint(s) 'sparse'.
    Best valid proposal 'DiscUnif' cannot take into account hint(s) 'sparse'.
    Best valid proposal 'DiscUnif' cannot take into account hint(s) 'sparse'.
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Best valid proposal 'Geometric' cannot take into account hint(s) 'sparse'.
    Best valid proposal 'Geometric' cannot take into account hint(s) 'sparse'.
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Starting contrastive divergence estimation via CD-MCMLE:
    Iteration 1 of at most 60:
    Convergence test P-value:5.1e-284
    The log-likelihood improved by 2.001.
    Iteration 2 of at most 60:
    Convergence test P-value:4.3e-156
    The log-likelihood improved by 1.112.
    Iteration 3 of at most 60:
    Convergence test P-value:3.3e-28
    The log-likelihood improved by 0.123.
    Iteration 4 of at most 60:
    Convergence test P-value:1.1e-08
    The log-likelihood improved by 0.03652.
    Iteration 5 of at most 60:
    Convergence test P-value:1.5e-01
    The log-likelihood improved by 0.006324.
    Iteration 6 of at most 60:
    Convergence test P-value:4.2e-02
    The log-likelihood improved by 0.009149.
    Iteration 7 of at most 60:
    Convergence test P-value:8.1e-01
    Convergence detected. Stopping.
    The log-likelihood improved by 0.001764.
    Finished CD.
    Starting Monte Carlo maximum likelihood estimation (MCMLE):
    Iteration 1 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.6088.
    Estimating equations are not within tolerance region.
    Iteration 2 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0397.
    Convergence test p-value: 0.1845. Not converged with 99% confidence; increasing sample size.
    Iteration 3 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0214.
    Convergence test p-value: 0.0327. Not converged with 99% confidence; increasing sample size.
    Iteration 4 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0043.
    Convergence test p-value: 0.0098. Converged with 99% confidence.
    Finished MCMLE.
    Evaluating log-likelihood at the estimate. Setting up bridge sampling...
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Using 16 bridges: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 .
    Note: Null model likelihood calculation is not implemented for valued
    ERGMs at this time. This means that all likelihood-based inference
    (LRT, Analysis of Deviance, AIC, BIC, etc.) is only valid between
    models with the same reference distribution and constraints.
    This model was fit using MCMC. To examine model diagnostics and check
    for degeneracy, use the mcmc.diagnostics() function.
    Starting contrastive divergence estimation via CD-MCMLE:
    Iteration 1 of at most 60:
    Convergence test P-value:5.7e-103
    The log-likelihood improved by 1.728.
    Iteration 2 of at most 60:
    Convergence test P-value:1.5e-47
    The log-likelihood improved by 0.5326.
    Iteration 3 of at most 60:
    Convergence test P-value:4.3e-22
    The log-likelihood improved by 0.2203.
    Iteration 4 of at most 60:
    Convergence test P-value:9.3e-16
    The log-likelihood improved by 0.1453.
    Iteration 5 of at most 60:
    Convergence test P-value:8.9e-08
    The log-likelihood improved by 0.06498.
    Iteration 6 of at most 60:
    Convergence test P-value:2.4e-04
    The log-likelihood improved by 0.03186.
    Iteration 7 of at most 60:
    Convergence test P-value:9.7e-03
    The log-likelihood improved by 0.01804.
    Iteration 8 of at most 60:
    Convergence test P-value:1.6e-03
    The log-likelihood improved by 0.02509.
    Iteration 9 of at most 60:
    Convergence test P-value:1.7e-02
    The log-likelihood improved by 0.01585.
    Iteration 10 of at most 60:
    Convergence test P-value:4.1e-01
    The log-likelihood improved by 0.00345.
    Iteration 11 of at most 60:
    Convergence test P-value:2.6e-02
    The log-likelihood improved by 0.01443.
    Iteration 12 of at most 60:
    Convergence test P-value:8.2e-03
    The log-likelihood improved by 0.01849.
    Iteration 13 of at most 60:
    Convergence test P-value:7.8e-01
    Convergence detected. Stopping.
    The log-likelihood improved by 0.001006.
    Finished CD.
    Starting Monte Carlo maximum likelihood estimation (MCMLE):
    Iteration 1 of at most 60:
    Optimizing with step length 0.5407.
    The log-likelihood improved by 3.2326.
    Estimating equations are not within tolerance region.
    Iteration 2 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 2.3468.
    Estimating equations are not within tolerance region.
    Iteration 3 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0038.
    Convergence test p-value: 0.0001. Converged with 99% confidence.
    Finished MCMLE.
    Evaluating log-likelihood at the estimate. Setting up bridge sampling...
    Using 16 bridges: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 .
    Note: Null model likelihood calculation is not implemented for valued
    ERGMs at this time. This means that all likelihood-based inference
    (LRT, Analysis of Deviance, AIC, BIC, etc.) is only valid between
    models with the same reference distribution and constraints.
    This model was fit using MCMC. To examine model diagnostics and check
    for degeneracy, use the mcmc.diagnostics() function.
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Starting contrastive divergence estimation via CD-MCMLE:
    Iteration 1 of at most 60:
    Convergence test P-value:0e+00
    The log-likelihood improved by 1.454.
    Iteration 2 of at most 60:
    Convergence test P-value:2e-185
    The log-likelihood improved by 1.153.
    Iteration 3 of at most 60:
    Convergence test P-value:1.2e-22
    The log-likelihood improved by 0.08417.
    Iteration 4 of at most 60:
    Convergence test P-value:2.4e-03
    The log-likelihood improved by 0.0135.
    Iteration 5 of at most 60:
    Convergence test P-value:5.7e-04
    The log-likelihood improved by 0.01593.
    Iteration 6 of at most 60:
    Convergence test P-value:8.7e-01
    Convergence detected. Stopping.
    The log-likelihood improved by 0.00163.
    Finished CD.
    Starting Monte Carlo maximum likelihood estimation (MCMLE):
    Iteration 1 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 1.6295.
    Estimating equations are not within tolerance region.
    Iteration 2 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.2131.
    Estimating equations are not within tolerance region.
    Iteration 3 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.2099.
    Estimating equations are not within tolerance region.
    Iteration 4 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.1440.
    Estimating equations are not within tolerance region.
    Iteration 5 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0646.
    Convergence test p-value: 0.7557. Not converged with 99% confidence; increasing sample size.
    Iteration 6 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.1043.
    Estimating equations are not within tolerance region.
    Estimating equations did not move closer to tolerance region more than 1 time(s) in 4 steps; increasing sample size.
    Iteration 7 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0309.
    Convergence test p-value: 0.0149. Not converged with 99% confidence; increasing sample size.
    Iteration 8 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0095.
    Convergence test p-value: 0.0021. Converged with 99% confidence.
    Finished MCMLE.
    Evaluating log-likelihood at the estimate. Setting up bridge sampling...
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Using 16 bridges: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 .
    Note: Null model likelihood calculation is not implemented for valued
    ERGMs at this time. This means that all likelihood-based inference
    (LRT, Analysis of Deviance, AIC, BIC, etc.) is only valid between
    models with the same reference distribution and constraints.
    This model was fit using MCMC. To examine model diagnostics and check
    for degeneracy, use the mcmc.diagnostics() function.
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Starting contrastive divergence estimation via CD-MCMLE:
    Iteration 1 of at most 60:
    Convergence test P-value:0e+00
    The log-likelihood improved by 1.126.
    Iteration 2 of at most 60:
    Convergence test P-value:9e-221
    The log-likelihood improved by 1.531.
    Iteration 3 of at most 60:
    Convergence test P-value:4e-49
    The log-likelihood improved by 0.2457.
    Iteration 4 of at most 60:
    Convergence test P-value:1.4e-04
    The log-likelihood improved by 0.02024.
    Iteration 5 of at most 60:
    Convergence test P-value:4.9e-02
    The log-likelihood improved by 0.008923.
    Iteration 6 of at most 60:
    Convergence test P-value:2.8e-01
    The log-likelihood improved by 0.005004.
    Iteration 7 of at most 60:
    Convergence test P-value:7.3e-01
    Convergence detected. Stopping.
    The log-likelihood improved by 0.002185.
    Finished CD.
    Starting Monte Carlo maximum likelihood estimation (MCMLE):
    Iteration 1 of at most 60:
    Optimizing with step length 0.7979.
    The log-likelihood improved by 2.1714.
    Estimating equations are not within tolerance region.
    Iteration 2 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.1799.
    Estimating equations are not within tolerance region.
    Iteration 3 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0094.
    Convergence test p-value: 0.7515. Not converged with 99% confidence; increasing sample size.
    Iteration 4 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0565.
    Convergence test p-value: 0.7302. Not converged with 99% confidence; increasing sample size.
    Iteration 5 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0473.
    Convergence test p-value: 0.1250. Not converged with 99% confidence; increasing sample size.
    Iteration 6 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0480.
    Convergence test p-value: 0.0307. Not converged with 99% confidence; increasing sample size.
    Iteration 7 of at most 60:
    Optimizing with step length 1.0000.
    The log-likelihood improved by 0.0180.
    Convergence test p-value: 0.0040. Converged with 99% confidence.
    Finished MCMLE.
    Evaluating log-likelihood at the estimate. Setting up bridge sampling...
    Best valid proposal 'Binomial' cannot take into account hint(s) 'sparse'.
    Using 16 bridges: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 .
    Note: Null model likelihood calculation is not implemented for valued
    ERGMs at this time. This means that all likelihood-based inference
    (LRT, Analysis of Deviance, AIC, BIC, etc.) is only valid between
    models with the same reference distribution and constraints.
    This model was fit using MCMC. To examine model diagnostics and check
    for degeneracy, use the mcmc.diagnostics() function.
    Evaluating network in model.
    Initializing unconstrained Metropolis-Hastings proposal: 'ergm.count:MH_PoissonTNT'.
    Initializing model...
    Model initialized.
    Using initial method 'CD'.
    Fitting initial model.
    Starting contrastive divergence estimation via CD-MCMLE:
    Iteration 1 of at most 60:
    Convergence test P-value:0e+00
    The log-likelihood improved by 1.383.
    Iteration 2 of at most 60:
    Convergence test P-value:0e+00
    The log-likelihood improved by 1.431.
    Iteration 3 of at most 60:
    Convergence test P-value:2.8e-180
    The log-likelihood improved by 0.7104.
    Iteration 4 of at most 60:
    Convergence test P-value:8.5e-26
    The log-likelihood improved by 0.07414.
    Iteration 5 of at most 60:
    Convergence test P-value:7.4e-05
    The log-likelihood improved by 0.01627.
    Iteration 6 of at most 60:
    Convergence test P-value:1.1e-01
    The log-likelihood improved by 0.006525.
    Iteration 7 of at most 60:
    Convergence test P-value:2.3e-01
    The log-likelihood improved by 0.005324.
    Iteration 8 of at most 60:
    Convergence test P-value:8.7e-03
    The log-likelihood improved by 0.01017.
    Iteration 9 of at most 60:
    Convergence test P-value:5.7e-02
    The log-likelihood improved by 0.007716.
    Iteration 10 of at most 60:
    Convergence test P-value:2.4e-01
    The log-likelihood improved by 0.005166.
    Iteration 11 of at most 60:
    Convergence test P-value:2.1e-01
    The log-likelihood improved by 0.005395.
    Iteration 12 of at most 60:
    Convergence test P-value:1.5e-01
    The log-likelihood improved by 0.005824.
    Iteration 13 of at most 60:
    Convergence test P-value:1.2e-01
    The log-likelihood improved by 0.006431.
    Iteration 14 of at most 60:
    Convergence test P-value:6.4e-01
    Convergence detected. Stopping.
    The log-likelihood improved by 0.003008.
    Finished CD.
    Starting Monte Carlo maximum likelihood estimation (MCMLE):
    Density guard set to 10000 from an initial count of 78 edges.
    
    Iteration 1 of at most 50 with parameter:
     sum nonzero
     1.14757556 -4.54002413
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -0.02256609 0.06934813
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.31689416 -1.06701157
     absdiff.sum.faction.id.4 nodecovar
     -1.21963250 3.16504720
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 512.
    Average estimating function values:
     sum nonzero
     521.159420 26.004831
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -46.845411 317.628019
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     86.574879 146.033816
     absdiff.sum.faction.id.4 nodecovar
     -9.507246 248.987576
    Starting MCMLE Optimization...
    Optimizing with step length 0.0383.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.7624.
    Estimating equations are not within tolerance region.
    
    Iteration 2 of at most 50 with parameter:
     sum nonzero
     1.34681434 -5.48331623
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.09059381 0.02167195
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.32433100 -1.07020557
     absdiff.sum.faction.id.4 nodecovar
     -0.56070256 2.58833824
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 256.
    Average estimating function values:
     sum nonzero
     489.315068 25.558904
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -44.487671 301.312329
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     82.347945 133.920548
     absdiff.sum.faction.id.4 nodecovar
     -8.284932 229.931123
    Starting MCMLE Optimization...
    Optimizing with step length 0.0649.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.9112.
    Distance from origin on tolerance region scale: 9062.8346363364 (previously 10333.3386740638).
    Estimating equations are not within tolerance region.
    
    Iteration 3 of at most 50 with parameter:
     sum nonzero
     1.43771346 -5.82475754
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.24262250 -0.06581077
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.41729149 -1.09081280
     absdiff.sum.faction.id.4 nodecovar
     -0.46961916 2.36915223
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 1024.
    Average estimating function values:
     sum nonzero
     633.22727 23.47727
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     573.08182 153.16136
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     73.91591 75.56591
     absdiff.sum.faction.id.4 nodecovar
     182.14773 288.08916
    Starting MCMLE Optimization...
    Optimizing with step length 0.0693.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.8659.
    Distance from origin on tolerance region scale: 7757.21280174457 (previously 7063.19528803493).
    Estimating equations are not within tolerance region.
    
    Iteration 4 of at most 50 with parameter:
     sum nonzero
     1.5343320 -6.4043232
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.1745123 -0.0870523
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.4300502 -0.9986735
     absdiff.sum.faction.id.4 nodecovar
     -0.4675286 2.1416028
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 512.
    Average estimating function values:
     sum nonzero
     581.84259 21.23765
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     524.96914 143.89815
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     67.48148 70.41975
     absdiff.sum.faction.id.4 nodecovar
     168.37346 266.28517
    Starting MCMLE Optimization...
    Optimizing with step length 0.0562.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.2744.
    Distance from origin on tolerance region scale: 8049.24321748246 (previously 9503.37183970382).
    Estimating equations are not within tolerance region.
    
    Iteration 5 of at most 50 with parameter:
     sum nonzero
     1.6085812 -6.8444890
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.1140250 -0.1358107
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.4472581 -0.9009664
     absdiff.sum.faction.id.4 nodecovar
     -0.4916166 1.9834600
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 1024.
    Average estimating function values:
     sum nonzero
     392.944444 -4.452381
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     516.166667 27.011905
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     38.317460 43.384921
     absdiff.sum.faction.id.4 nodecovar
     164.392857 213.799477
    Starting MCMLE Optimization...
    Optimizing with step length 0.0813.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.9103.
    Distance from origin on tolerance region scale: 5758.42361323797 (previously 23012.4750823145).
    Estimating equations are not within tolerance region.
    
    Iteration 6 of at most 50 with parameter:
     sum nonzero
     1.76127416 -7.50168807
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -0.06084443 -0.11936371
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.46319424 -0.90553456
     absdiff.sum.faction.id.4 nodecovar
     -0.49770339 1.89788183
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 512.
    Average estimating function values:
     sum nonzero
     364.816393 -3.668852
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     469.209836 28.078689
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     34.055738 36.288525
     absdiff.sum.faction.id.4 nodecovar
     151.321311 192.085789
    Starting MCMLE Optimization...
    Optimizing with step length 0.1007.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 2.0120.
    Distance from origin on tolerance region scale: 3954.60704358574 (previously 4760.06349209778).
    Estimating equations are not within tolerance region.
    
    Iteration 7 of at most 50 with parameter:
     sum nonzero
     1.8653686 -7.7669190
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -0.3388983 -0.1459032
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.5051400 -0.8829183
     absdiff.sum.faction.id.4 nodecovar
     -0.5188820 2.1900313
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 512.
    Average estimating function values:
     sum nonzero
     336.466049 -2.274691
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     426.956790 17.833333
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     29.753086 35.188272
     absdiff.sum.faction.id.4 nodecovar
     140.466049 175.760734
    Starting MCMLE Optimization...
    Optimizing with step length 0.1056.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 2.1527.
    Distance from origin on tolerance region scale: 3850.29914671729 (previously 4623.56751579031).
    Estimating equations are not within tolerance region.
    
    Iteration 8 of at most 50 with parameter:
     sum nonzero
     1.92683426 -7.92485322
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -0.59255363 -0.08492026
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.46717107 -0.85929250
     absdiff.sum.faction.id.4 nodecovar
     -0.54576801 2.45863178
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 1024.
    Average estimating function values:
     sum nonzero
     962.34773 20.72955
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -38.98636 104.97955
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     87.55455 116.00227
     absdiff.sum.faction.id.4 nodecovar
     291.38636 425.06811
    Starting MCMLE Optimization...
    Optimizing with step length 0.0391.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 2.5770.
    Distance from origin on tolerance region scale: 33710.4490246565 (previously 52900.2504868789).
    Estimating equations are not within tolerance region.
    
    Iteration 9 of at most 50 with parameter:
     sum nonzero
     2.17421495 -10.17109988
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -0.40887326 -0.08741896
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.43771947 -0.77404037
     absdiff.sum.faction.id.4 nodecovar
     -0.51035544 1.66151655
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 512.
    Average estimating function values:
     sum nonzero
     933.49885 20.81755
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -36.00000 102.86605
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     85.03695 110.54503
     absdiff.sum.faction.id.4 nodecovar
     281.80370 403.65381
    Starting MCMLE Optimization...
    Optimizing with step length 0.0780.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 3.1036.
    Distance from origin on tolerance region scale: 10177.410440676 (previously 11381.2102337954).
    Estimating equations are not within tolerance region.
    
    Iteration 10 of at most 50 with parameter:
     sum nonzero
     2.1862675 -10.4576029
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -0.2366519 -0.0667600
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.3737137 -0.6826825
     absdiff.sum.faction.id.4 nodecovar
     -0.5148818 1.4260390
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 1024.
    Average estimating function values:
     sum nonzero
     852.00549 19.59615
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -37.73077 101.48077
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     81.28846 92.17033
     absdiff.sum.faction.id.4 nodecovar
     246.52473 350.37316
    Starting MCMLE Optimization...
    Optimizing with step length 0.0716.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.6966.
    Distance from origin on tolerance region scale: 6595.59423028242 (previously 8021.08916646425).
    Estimating equations are not within tolerance region.
    
    Iteration 11 of at most 50 with parameter:
     sum nonzero
     2.15742949 -10.48290163
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -0.12967638 -0.06646713
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.36010755 -0.61477014
     absdiff.sum.faction.id.4 nodecovar
     -0.51442340 1.35246212
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 4096.
    Average estimating function values:
     sum nonzero
     549.742604 -6.470414
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -46.636095 51.100592
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     52.479290 71.408284
     absdiff.sum.faction.id.4 nodecovar
     170.840237 263.974424
    Starting MCMLE Optimization...
    Optimizing with step length 0.0844.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.9243.
    Distance from origin on tolerance region scale: 5384.13238783937 (previously 7939.76908255329).
    Estimating equations are not within tolerance region.
    
    Iteration 12 of at most 50 with parameter:
     sum nonzero
     2.09145129 -9.93039321
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -0.03142636 -0.06650590
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.36626784 -0.61689381
     absdiff.sum.faction.id.4 nodecovar
     -0.53257023 1.36161694
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 2048.
    Average estimating function values:
     sum nonzero
     494.338164 -6.570048
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -45.309179 44.980676
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     47.198068 61.545894
     absdiff.sum.faction.id.4 nodecovar
     152.642512 235.594748
    Starting MCMLE Optimization...
    Optimizing with step length 0.0728.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.1836.
    Distance from origin on tolerance region scale: 4443.33772035737 (previously 5272.72374701386).
    Estimating equations are not within tolerance region.
    
    Iteration 13 of at most 50 with parameter:
     sum nonzero
     2.02571025 -9.45016396
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.03271826 -0.05932386
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.38193436 -0.60197374
     absdiff.sum.faction.id.4 nodecovar
     -0.56170635 1.40404017
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 1024.
    Average estimating function values:
     sum nonzero
     456.610315 -5.621777
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     -35.088825 43.249284
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     42.939828 57.805158
     absdiff.sum.faction.id.4 nodecovar
     131.905444 212.702325
    Starting MCMLE Optimization...
    Optimizing with step length 0.0987.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.9845.
    Distance from origin on tolerance region scale: 4064.79388309244 (previously 4952.58613488379).
    Estimating equations are not within tolerance region.
    
    Iteration 14 of at most 50 with parameter:
     sum nonzero
     1.93464053 -8.74069515
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.10537371 -0.07003491
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.39003199 -0.57699127
     absdiff.sum.faction.id.4 nodecovar
     -0.57883999 1.44881609
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 2048.
    Average estimating function values:
     sum nonzero
     464.456818 -3.056818
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     283.822727 41.725000
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     43.475000 63.822727
     absdiff.sum.faction.id.4 nodecovar
     138.825000 220.361898
    Starting MCMLE Optimization...
    Optimizing with step length 0.1198.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 2.0748.
    Distance from origin on tolerance region scale: 2883.82880307752 (previously 5696.563373274).
    Estimating equations are not within tolerance region.
    
    Iteration 15 of at most 50 with parameter:
     sum nonzero
     1.87518335 -8.32375781
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.09498195 -0.07615420
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.39582574 -0.56654231
     absdiff.sum.faction.id.4 nodecovar
     -0.58367287 1.43846114
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 1024.
    Average estimating function values:
     sum nonzero
     414.672165 -2.847423
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     249.573196 32.964948
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     38.764948 59.868041
     absdiff.sum.faction.id.4 nodecovar
     126.348454 200.655604
    Starting MCMLE Optimization...
    Optimizing with step length 0.1568.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 3.2153.
    Distance from origin on tolerance region scale: 2608.68735455347 (previously 3258.92852265549).
    Estimating equations are not within tolerance region.
    
    Iteration 16 of at most 50 with parameter:
     sum nonzero
     1.77488196 -7.78762536
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.08710843 -0.04411953
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.39648873 -0.56767866
     absdiff.sum.faction.id.4 nodecovar
     -0.59332614 1.46009494
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 1024.
    Average estimating function values:
     sum nonzero
     181.25926 -23.81481
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     225.78788 14.48485
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     17.45791 27.03367
     absdiff.sum.faction.id.4 nodecovar
     63.53535 103.49873
    Starting MCMLE Optimization...
    Optimizing with step length 0.1049.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 1.7832.
    Distance from origin on tolerance region scale: 3227.26906855036 (previously 9579.99274657682).
    Estimating equations are not within tolerance region.
    
    Iteration 17 of at most 50 with parameter:
     sum nonzero
     1.66249017 -7.04439811
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.10217278 -0.06877268
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.41513175 -0.54217634
     absdiff.sum.faction.id.4 nodecovar
     -0.61958629 1.48278696
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 262144.
    Average estimating function values:
     sum nonzero
     291.926641 -4.104247
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     366.664093 27.938224
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     28.181467 41.324324
     absdiff.sum.faction.id.4 nodecovar
     101.413127 146.530061
    Starting MCMLE Optimization...
    Optimizing with step length 0.1822.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 2.1452.
    Distance from origin on tolerance region scale: 1287.88504271744 (previously 1758.86296646802).
    Estimating equations are not within tolerance region.
    
    Iteration 18 of at most 50 with parameter:
     sum nonzero
     1.56381331 -6.42136263
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.10341604 -0.07258483
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.43872666 -0.55666662
     absdiff.sum.faction.id.4 nodecovar
     -0.63375292 1.55214143
    Starting unconstrained MCMC...
    Back from unconstrained MCMC.
    New interval = 262144.
    Average estimating function values:
     sum nonzero
     285.50133 2.35200
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     355.91467 28.37333
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     27.36267 40.57333
     absdiff.sum.faction.id.4 nodecovar
     100.07200 140.74773
    Starting MCMLE Optimization...
    Optimizing with step length 0.2152.
    Using lognormal metric (see control.ergm function).
    Using log-normal approx (no optim)
    The log-likelihood improved by 2.4121.
    Distance from origin on tolerance region scale: 1039.38715545015 (previously 1403.00000659055).
    Estimating equations are not within tolerance region.
    
    Iteration 19 of at most 50 with parameter:
     sum nonzero
     1.49135894 -5.91164668
    nodefactor.sum.leader.TRUE absdiff.sum.faction.id.1
     0.04436801 -0.07805467
     absdiff.sum.faction.id.2 absdiff.sum.faction.id.3
     -0.44924758 -0.60235655
     absdiff.sum.faction.id.4 nodecovar
     -0.66879138 1.72183431
    Starting unconstrained MCMC...
    Killed
Flavor: r-devel-linux-x86_64-debian-clang