library(readr)
library(igraph)
library(ergm)
library(network)
library(intergraph)
library(parallel)
library(latticeExtra)
panel_report_meta <- read_csv(url("https://raw.githubusercontent.com/chanyaaaa/WTO_panel_citation/master/panel_reports_meta.csv"))
panel_citation <- as.matrix(read_csv(url("https://raw.githubusercontent.com/chanyaaaa/WTO_panel_citation/master/panel_citation.csv"),
col_names = F))
colnames(panel_citation) <- panel_report_meta$dispute
rownames(panel_citation) <- panel_report_meta$dispute
shared_panels <- as.matrix(read_csv(url("https://raw.githubusercontent.com/chanyaaaa/WTO_panel_citation/master/shared_panels.csv"),
col_names = F))
colnames(shared_panels) <- panel_report_meta$dispute
rownames(shared_panels) <- panel_report_meta$dispute
shared_personnel <- as.matrix(read_csv(url("https://raw.githubusercontent.com/chanyaaaa/WTO_panel_citation/master/shared_personnel.csv"),
col_names = F))
colnames(shared_personnel) <- panel_report_meta$dispute
rownames(shared_personnel) <- panel_report_meta$dispute
wto_panel_net <- graph_from_adjacency_matrix(panel_citation, diag = F)
set.seed(123)
plot(wto_panel_net,
vertex.size = 5,
vertex.color = "red",
vertex.label = V(wto_panel_net)$name,
vertex.label.cex = .5,
vertex.frame.color = NA,
edge.arrow.size = 0.1,
layout = layout.fruchterman.reingold,
)
shared_panels_net <- graph_from_adjacency_matrix(shared_panels, mode = "undirected", diag = F, weighted = T)
set.seed(123)
plot(shared_panels_net,
vertex.size = 5,
vertex.color = "red",
vertex.label = V(shared_panels_net)$name,
vertex.label.cex = .5,
vertex.frame.color = NA,
edge.width = E(shared_panels_net)$weight,
layout = layout.fruchterman.reingold,
)
shared_personnel_net <- graph_from_adjacency_matrix(shared_personnel, mode = "undirected", diag = F, weighted = T)
set.seed(123)
plot(shared_personnel_net,
vertex.size = 5,
vertex.color = "red",
vertex.label = V(shared_personnel_net)$name,
vertex.label.cex = .5,
vertex.frame.color = NA,
edge.width = E(shared_personnel_net)$weight,
layout = layout.fruchterman.reingold,
)
wto_panel_net <- asNetwork(wto_panel_net)
wto_panel_net %e% "shared_panel" <- shared_panels
wto_panel_net %e% "shared_personnel" <- shared_personnel
set.seed(123)
wto_panel_model.01 <- ergm(wto_panel_net ~ edges + gwesp(0.2, fixed = T),
control=control.ergm(parallel= detectCores() -1, parallel.type="PSOCK"))
summary(wto_panel_model.01)
## Call:
## ergm(formula = wto_panel_net ~ edges + gwesp(0.2, fixed = T),
## control = control.ergm(parallel = detectCores() - 1, parallel.type = "PSOCK"))
##
## Monte Carlo Maximum Likelihood Results:
##
## Estimate Std. Error MCMC % z value Pr(>|z|)
## edges -3.72932 0.08107 0 -46.00 <1e-04 ***
## gwesp.fixed.0.2 0.92630 0.05891 0 15.72 <1e-04 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Null Deviance: 16622 on 11990 degrees of freedom
## Residual Deviance: 6415 on 11988 degrees of freedom
##
## AIC: 6419 BIC: 6434 (Smaller is better. MC Std. Err. = 0.8602)
set.seed(123)
wto_panel_model.02 <- ergm(wto_panel_net ~ edges + gwesp(0.2, fixed = T) + edgecov(shared_panels),
control=control.ergm(parallel= detectCores() -1, parallel.type="PSOCK"))
summary(wto_panel_model.02)
## Call:
## ergm(formula = wto_panel_net ~ edges + gwesp(0.2, fixed = T) +
## edgecov(shared_panels), control = control.ergm(parallel = detectCores() -
## 1, parallel.type = "PSOCK"))
##
## Monte Carlo Maximum Likelihood Results:
##
## Estimate Std. Error MCMC % z value Pr(>|z|)
## edges -3.74855 0.07900 0 -47.452 <1e-04 ***
## gwesp.fixed.0.2 0.91316 0.05648 0 16.169 <1e-04 ***
## edgecov.shared_panels 0.58843 0.08420 0 6.989 <1e-04 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Null Deviance: 16622 on 11990 degrees of freedom
## Residual Deviance: 6377 on 11987 degrees of freedom
##
## AIC: 6383 BIC: 6406 (Smaller is better. MC Std. Err. = 0.6602)
set.seed(123)
wto_panel_model.03 <- ergm(wto_panel_net ~edges + gwesp(0.2, fixed = T) + edgecov(shared_panels) + edgecov(shared_personnel),
control=control.ergm(parallel= detectCores() -1, parallel.type="PSOCK"))
summary(wto_panel_model.03)
## Call:
## ergm(formula = wto_panel_net ~ edges + gwesp(0.2, fixed = T) +
## edgecov(shared_panels) + edgecov(shared_personnel), control = control.ergm(parallel = detectCores() -
## 1, parallel.type = "PSOCK"))
##
## Monte Carlo Maximum Likelihood Results:
##
## Estimate Std. Error MCMC % z value Pr(>|z|)
## edges -3.69239 0.10755 0 -34.330 <1e-04 ***
## gwesp.fixed.0.2 0.91313 0.06372 0 14.330 <1e-04 ***
## edgecov.shared_panels 0.60052 0.08740 0 6.871 <1e-04 ***
## edgecov.shared_personnel -0.02235 0.02463 0 -0.907 0.364
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Null Deviance: 16622 on 11990 degrees of freedom
## Residual Deviance: 6377 on 11986 degrees of freedom
##
## AIC: 6385 BIC: 6415 (Smaller is better. MC Std. Err. = 0.8424)
wto_panel_model.01.gof <- gof(wto_panel_model.03, GOF = ~ model + triadcensus + odegree + idegree + espartners + distance)
wto_panel_model.02.gof <- gof(wto_panel_model.03, GOF = ~ model + triadcensus + odegree + idegree + espartners + distance)
wto_panel_model.03.gof <- gof(wto_panel_model.03, GOF = ~ model + triadcensus + odegree + idegree + espartners + distance)
plot(wto_panel_model.01.gof)
plot(wto_panel_model.02.gof)
plot(wto_panel_model.03.gof)
mcmc.diagnostics(wto_panel_model.01)
## Sample statistics summary:
##
## Iterations = 139264:827392
## Thinning interval = 8192
## Number of chains = 7
## Sample size per chain = 85
##
## 1. Empirical mean and standard deviation for each variable,
## plus standard error of the mean:
##
## Mean SD Naive SE Time-series SE
## edges -4.588 71.76 2.942 6.564
## gwesp.fixed.0.2 -6.523 98.74 4.048 8.856
##
## 2. Quantiles for each variable:
##
## 2.5% 25% 50% 75% 97.5%
## edges -166.2 -45.50 -4.000 42.50 130.1
## gwesp.fixed.0.2 -226.2 -64.31 -4.995 59.91 179.7
##
##
## Are sample statistics significantly different from observed?
## edges gwesp.fixed.0.2 Overall (Chi^2)
## diff. -4.5882353 -6.5230726 NA
## test stat. -0.7670676 -0.8116659 0.7387144
## P-val. 0.4430413 0.4169834 0.6924118
##
## Sample statistics cross-correlations:
## edges gwesp.fixed.0.2
## edges 1.0000000 0.9850634
## gwesp.fixed.0.2 0.9850634 1.0000000
##
## Sample statistics auto-correlation:
## Chain 1
## edges gwesp.fixed.0.2
## Lag 0 1.000000000 1.000000000
## Lag 8192 0.541044066 0.549457689
## Lag 16384 0.190134518 0.229973331
## Lag 24576 0.044892469 0.061353619
## Lag 32768 -0.007505901 0.003766333
## Lag 40960 -0.026195327 -0.030599548
## Chain 2
## edges gwesp.fixed.0.2
## Lag 0 1.00000000 1.00000000
## Lag 8192 0.64311484 0.63950033
## Lag 16384 0.44386737 0.45771196
## Lag 24576 0.20070946 0.22392535
## Lag 32768 0.02940063 0.05306157
## Lag 40960 -0.12242303 -0.11182385
## Chain 3
## edges gwesp.fixed.0.2
## Lag 0 1.0000000 1.0000000
## Lag 8192 0.5436563 0.4968077
## Lag 16384 0.3012825 0.2593121
## Lag 24576 0.2834614 0.2624314
## Lag 32768 0.2272395 0.2107273
## Lag 40960 0.1372958 0.1348736
## Chain 4
## edges gwesp.fixed.0.2
## Lag 0 1.00000000 1.00000000
## Lag 8192 0.69435561 0.69426770
## Lag 16384 0.47187112 0.45995140
## Lag 24576 0.29632041 0.30365677
## Lag 32768 0.15328748 0.14260939
## Lag 40960 0.02432548 0.04892741
## Chain 5
## edges gwesp.fixed.0.2
## Lag 0 1.00000000 1.00000000
## Lag 8192 0.49270884 0.45845691
## Lag 16384 0.20745993 0.14443528
## Lag 24576 0.15811751 0.14430274
## Lag 32768 0.05424337 0.03005498
## Lag 40960 -0.03833745 -0.05308058
## Chain 6
## edges gwesp.fixed.0.2
## Lag 0 1.00000000 1.00000000
## Lag 8192 0.59007101 0.57194822
## Lag 16384 0.29073569 0.30825233
## Lag 24576 0.16987113 0.18572464
## Lag 32768 0.11052792 0.10411776
## Lag 40960 -0.05117164 -0.06799422
## Chain 7
## edges gwesp.fixed.0.2
## Lag 0 1.0000000 1.0000000
## Lag 8192 0.6684125 0.6586543
## Lag 16384 0.5692664 0.5656573
## Lag 24576 0.4844374 0.4780437
## Lag 32768 0.3377173 0.3384555
## Lag 40960 0.2295567 0.2256378
##
## Sample statistics burn-in diagnostic (Geweke):
## Chain 1
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2
## 0.2094 0.3210
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2
## 0.8341221 0.7482092
## Joint P-value (lower = worse): 0.2891277 .
## Chain 2
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2
## 0.8895 0.7941
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2
## 0.3737459 0.4271431
## Joint P-value (lower = worse): 0.7828912 .
## Chain 3
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2
## 2.003 1.996
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2
## 0.04514348 0.04591024
## Joint P-value (lower = worse): 0.4777455 .
## Chain 4
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2
## -0.1693 -0.1324
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2
## 0.8655871 0.8946637
## Joint P-value (lower = worse): 0.9809669 .
## Chain 5
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2
## 0.1698 0.1495
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2
## 0.8651628 0.8811908
## Joint P-value (lower = worse): 0.9917481 .
## Chain 6
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2
## 0.9322 0.9342
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2
## 0.3512105 0.3502172
## Joint P-value (lower = worse): 0.8383678 .
## Chain 7
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2
## -1.036 -1.105
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2
## 0.2999820 0.2691882
## Joint P-value (lower = worse): 0.7781036 .
##
## MCMC diagnostics shown here are from the last round of simulation, prior to computation of final parameter estimates. Because the final estimates are refinements of those used for this simulation run, these diagnostics may understate model performance. To directly assess the performance of the final model on in-model statistics, please use the GOF command: gof(ergmFitObject, GOF=~model).
mcmc.diagnostics(wto_panel_model.02)
## Sample statistics summary:
##
## Iterations = 294912:1474560
## Thinning interval = 16384
## Number of chains = 7
## Sample size per chain = 73
##
## 1. Empirical mean and standard deviation for each variable,
## plus standard error of the mean:
##
## Mean SD Naive SE Time-series SE
## edges -10.4697 64.00 2.8311 4.094
## gwesp.fixed.0.2 -14.1948 89.52 3.9603 5.592
## edgecov.shared_panels -0.1135 12.92 0.5718 0.759
##
## 2. Quantiles for each variable:
##
## 2.5% 25% 50% 75% 97.5%
## edges -131.75 -55.00 -11.00 36.50 106.2
## gwesp.fixed.0.2 -179.44 -77.12 -12.71 49.92 154.7
## edgecov.shared_panels -24.25 -8.00 0.00 8.50 25.0
##
##
## Are sample statistics significantly different from observed?
## edges gwesp.fixed.0.2 edgecov.shared_panels Overall (Chi^2)
## diff. -10.469667319 -14.194826250 -0.1135029 NA
## test stat. -2.765453497 -2.711499766 -0.1575029 8.08904944
## P-val. 0.005684371 0.006697959 0.8748485 0.04687236
##
## Sample statistics cross-correlations:
## edges gwesp.fixed.0.2 edgecov.shared_panels
## edges 1.0000000 0.9801268 0.3868940
## gwesp.fixed.0.2 0.9801268 1.0000000 0.3872259
## edgecov.shared_panels 0.3868940 0.3872259 1.0000000
##
## Sample statistics auto-correlation:
## Chain 1
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.000000000 1.000000000 1.00000000
## Lag 16384 0.205252525 0.195756140 0.26358019
## Lag 32768 0.002882614 0.002980485 0.17748011
## Lag 49152 -0.033809227 -0.048724564 0.06374397
## Lag 65536 -0.045212140 -0.047675988 0.06683707
## Lag 81920 -0.056770697 -0.031908984 0.09368365
## Chain 2
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.0000000 1.00000000 1.00000000
## Lag 16384 0.2844351 0.24138211 0.32060115
## Lag 32768 -0.1219451 -0.15106481 -0.11820335
## Lag 49152 -0.1866672 -0.19288560 -0.13394792
## Lag 65536 -0.2511173 -0.19828634 -0.14818703
## Lag 81920 0.1186569 0.07570948 0.06015265
## Chain 3
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.000000000 1.000000000 1.000000000
## Lag 16384 0.530490170 0.509906943 0.403510767
## Lag 32768 0.288882726 0.308694917 0.260379808
## Lag 49152 0.217239228 0.215636234 0.038249832
## Lag 65536 0.039207164 0.055235191 0.009207241
## Lag 81920 -0.003676279 -0.002783644 0.024232005
## Chain 4
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.00000000 1.00000000 1.00000000
## Lag 16384 0.29644409 0.32063565 0.09166636
## Lag 32768 0.09078477 0.11157622 0.03378808
## Lag 49152 -0.05825371 -0.07540709 0.02463202
## Lag 65536 -0.15335553 -0.23200868 -0.09875086
## Lag 81920 -0.22821537 -0.24356435 0.11081031
## Chain 5
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.00000000 1.00000000 1.00000000
## Lag 16384 0.39552372 0.36427039 0.24656924
## Lag 32768 0.09559802 0.10238711 -0.02622055
## Lag 49152 0.09686949 0.07355758 0.24222613
## Lag 65536 -0.02656633 -0.03001282 0.02982365
## Lag 81920 -0.10793027 -0.11616079 -0.15195168
## Chain 6
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.000000000 1.000000000 1.000000000
## Lag 16384 0.235237944 0.207798635 0.213259923
## Lag 32768 0.004245257 -0.006122916 -0.070123442
## Lag 49152 0.016934973 -0.037071845 0.064240694
## Lag 65536 0.068672104 0.035493017 0.106152399
## Lag 81920 0.002569283 -0.003992582 0.003357262
## Chain 7
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.000000000 1.000000000 1.000000000
## Lag 16384 0.335917035 0.349401117 0.139631642
## Lag 32768 0.045849453 0.080777733 0.002936689
## Lag 49152 -0.086124586 -0.118546384 0.347607354
## Lag 65536 -0.002520493 -0.048821896 -0.099513606
## Lag 81920 0.037424354 -0.001261064 -0.199145105
##
## Sample statistics burn-in diagnostic (Geweke):
## Chain 1
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 1.682 1.943 4.553
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 9.256106e-02 5.201149e-02 5.298422e-06
## Joint P-value (lower = worse): 0.0126946 .
## Chain 2
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## -0.30292 0.07894 0.86545
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.7619497 0.9370776 0.3867944
## Joint P-value (lower = worse): 0.3269362 .
## Chain 3
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## -0.3763 -0.4570 -1.1795
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.7066913 0.6476976 0.2381955
## Joint P-value (lower = worse): 0.8560017 .
## Chain 4
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.5246 0.6987 2.7846
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.599855025 0.484710566 0.005358579
## Joint P-value (lower = worse): 0.2908666 .
## Chain 5
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 3.2614 3.7184 0.3737
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.0011084968 0.0002005107 0.7086261283
## Joint P-value (lower = worse): 0.1108889 .
## Chain 6
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.7428 0.5926 1.1197
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.4575762 0.5534642 0.2628571
## Joint P-value (lower = worse): 0.759567 .
## Chain 7
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## -0.8995 -0.6495 -2.0365
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.36838564 0.51603412 0.04170309
## Joint P-value (lower = worse): 0.1451812 .
##
## MCMC diagnostics shown here are from the last round of simulation, prior to computation of final parameter estimates. Because the final estimates are refinements of those used for this simulation run, these diagnostics may understate model performance. To directly assess the performance of the final model on in-model statistics, please use the GOF command: gof(ergmFitObject, GOF=~model).
mcmc.diagnostics(wto_panel_model.03)
## Sample statistics summary:
##
## Iterations = 155648:974848
## Thinning interval = 8192
## Number of chains = 7
## Sample size per chain = 101
##
## 1. Empirical mean and standard deviation for each variable,
## plus standard error of the mean:
##
## Mean SD Naive SE Time-series SE
## edges -3.8246 64.02 2.4078 4.2905
## gwesp.fixed.0.2 -4.6288 87.73 3.2993 5.6739
## edgecov.shared_panels -0.7072 13.20 0.4964 0.8153
## edgecov.shared_personnel -8.3946 168.06 6.3206 11.0066
##
## 2. Quantiles for each variable:
##
## 2.5% 25% 50% 75% 97.5%
## edges -135.3 -45.50 -4.000 42.00 123.7
## gwesp.fixed.0.2 -188.5 -60.13 -2.447 53.92 169.1
## edgecov.shared_panels -26.0 -10.00 0.000 9.00 23.0
## edgecov.shared_personnel -362.9 -119.00 -8.000 109.50 313.3
##
##
## Are sample statistics significantly different from observed?
## edges gwesp.fixed.0.2 edgecov.shared_panels
## diff. -3.8246110 -4.6288288 -0.7072136
## test stat. -0.8825833 -0.8157770 -0.9235597
## P-val. 0.3774614 0.4146277 0.3557156
## edgecov.shared_personnel Overall (Chi^2)
## diff. -8.3946252 NA
## test stat. -0.7532532 2.296086
## P-val. 0.4512978 0.685173
##
## Sample statistics cross-correlations:
## edges gwesp.fixed.0.2 edgecov.shared_panels
## edges 1.0000000 0.9835699 0.4653847
## gwesp.fixed.0.2 0.9835699 1.0000000 0.4769071
## edgecov.shared_panels 0.4653847 0.4769071 1.0000000
## edgecov.shared_personnel 0.9695559 0.9532862 0.4835112
## edgecov.shared_personnel
## edges 0.9695559
## gwesp.fixed.0.2 0.9532862
## edgecov.shared_panels 0.4835112
## edgecov.shared_personnel 1.0000000
##
## Sample statistics auto-correlation:
## Chain 1
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.00000000 1.00000000 1.0000000000
## Lag 8192 0.45046333 0.43651950 0.4042510149
## Lag 16384 0.18673140 0.14147727 0.1534430361
## Lag 24576 0.13361100 0.08050757 0.0491890914
## Lag 32768 0.11737395 0.08494353 -0.0023869086
## Lag 40960 0.09428062 0.05847478 -0.0003399683
## edgecov.shared_personnel
## Lag 0 1.00000000
## Lag 8192 0.45086118
## Lag 16384 0.18700751
## Lag 24576 0.13549750
## Lag 32768 0.05399027
## Lag 40960 0.03039310
## Chain 2
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.00000000 1.0000000 1.0000000
## Lag 8192 0.48997508 0.5077682 0.3845428
## Lag 16384 0.17318350 0.1799506 0.2090845
## Lag 24576 0.13603779 0.1550451 0.2550886
## Lag 32768 -0.02938998 0.0166963 0.1742157
## Lag 40960 -0.12066929 -0.0715848 0.1326253
## edgecov.shared_personnel
## Lag 0 1.00000000
## Lag 8192 0.44498949
## Lag 16384 0.14242151
## Lag 24576 0.10081595
## Lag 32768 -0.07145916
## Lag 40960 -0.10255518
## Chain 3
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.00000000 1.00000000 1.00000000
## Lag 8192 0.51494456 0.48157714 0.17377170
## Lag 16384 0.18808280 0.11097562 0.03502230
## Lag 24576 0.10908364 0.03720395 0.00913528
## Lag 32768 0.08139201 0.05699899 0.15519057
## Lag 40960 0.08824836 0.07053682 0.02723564
## edgecov.shared_personnel
## Lag 0 1.00000000
## Lag 8192 0.50850255
## Lag 16384 0.16042761
## Lag 24576 0.11400801
## Lag 32768 0.06599421
## Lag 40960 0.04768946
## Chain 4
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.0000000 1.0000000 1.0000000
## Lag 8192 0.5592154 0.5460224 0.4929922
## Lag 16384 0.3731059 0.3813799 0.2590899
## Lag 24576 0.2122358 0.2369954 0.2861028
## Lag 32768 0.1308311 0.1278726 0.2458864
## Lag 40960 0.1895495 0.1990546 0.1434693
## edgecov.shared_personnel
## Lag 0 1.0000000
## Lag 8192 0.5547523
## Lag 16384 0.3656697
## Lag 24576 0.2013203
## Lag 32768 0.1169880
## Lag 40960 0.1774161
## Chain 5
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.0000000 1.00000000 1.00000000
## Lag 8192 0.5089471 0.48221447 0.49637483
## Lag 16384 0.2950437 0.29698758 0.38303955
## Lag 24576 0.1167389 0.07688077 0.26488483
## Lag 32768 -0.1029901 -0.10272941 0.13848068
## Lag 40960 -0.1490851 -0.13552491 0.08104964
## edgecov.shared_personnel
## Lag 0 1.00000000
## Lag 8192 0.53947600
## Lag 16384 0.30262298
## Lag 24576 0.07609620
## Lag 32768 -0.09539039
## Lag 40960 -0.12867939
## Chain 6
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.00000000 1.00000000 1.00000000
## Lag 8192 0.51318003 0.49721871 0.37548759
## Lag 16384 0.34719580 0.30603285 0.22432604
## Lag 24576 0.27370857 0.23564245 0.15047362
## Lag 32768 0.20768711 0.16095552 0.03031621
## Lag 40960 0.02847814 0.02116772 -0.07452968
## edgecov.shared_personnel
## Lag 0 1.00000000
## Lag 8192 0.52163746
## Lag 16384 0.28792035
## Lag 24576 0.26111440
## Lag 32768 0.22951394
## Lag 40960 0.07231841
## Chain 7
## edges gwesp.fixed.0.2 edgecov.shared_panels
## Lag 0 1.00000000 1.00000000 1.000000000
## Lag 8192 0.39533115 0.36582838 0.395595959
## Lag 16384 0.31021990 0.26339001 0.144900942
## Lag 24576 0.30441400 0.31910199 0.082555279
## Lag 32768 0.21995893 0.21101017 0.105847476
## Lag 40960 0.03929676 0.03706958 0.002499552
## edgecov.shared_personnel
## Lag 0 1.00000000
## Lag 8192 0.42112295
## Lag 16384 0.31715208
## Lag 24576 0.28743662
## Lag 32768 0.20489700
## Lag 40960 0.02764382
##
## Sample statistics burn-in diagnostic (Geweke):
## Chain 1
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## -0.9502 -1.5165 -0.9210
## edgecov.shared_personnel
## -1.1301
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.3420290 0.1293968 0.3570509
## edgecov.shared_personnel
## 0.2584349
## Joint P-value (lower = worse): 0.09366804 .
## Chain 2
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.9256 1.2796 1.0521
## edgecov.shared_personnel
## 0.4322
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.3546576 0.2006927 0.2927509
## edgecov.shared_personnel
## 0.6656276
## Joint P-value (lower = worse): 0.2470874 .
## Chain 3
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## -1.150 -1.120 -2.422
## edgecov.shared_personnel
## -1.159
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.25030159 0.26263714 0.01543463
## edgecov.shared_personnel
## 0.24648028
## Joint P-value (lower = worse): 0.1938829 .
## Chain 4
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.7404 0.7734 1.9494
## edgecov.shared_personnel
## 0.9355
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.45908758 0.43930514 0.05124346
## edgecov.shared_personnel
## 0.34954346
## Joint P-value (lower = worse): 0.4667118 .
## Chain 5
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## -1.239 -1.423 -1.485
## edgecov.shared_personnel
## -1.219
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.2152052 0.1547094 0.1374875
## edgecov.shared_personnel
## 0.2228180
## Joint P-value (lower = worse): 0.5529701 .
## Chain 6
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## -0.3238 -0.2144 -0.6994
## edgecov.shared_personnel
## -0.5073
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.7460760 0.8302637 0.4842859
## edgecov.shared_personnel
## 0.6119730
## Joint P-value (lower = worse): 0.703946 .
## Chain 7
##
## Fraction in 1st window = 0.1
## Fraction in 2nd window = 0.5
##
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.1502 0.3673 0.1289
## edgecov.shared_personnel
## 0.7233
##
## Individual P-values (lower = worse):
## edges gwesp.fixed.0.2 edgecov.shared_panels
## 0.8805831 0.7133709 0.8974205
## edgecov.shared_personnel
## 0.4694706
## Joint P-value (lower = worse): 0.4576132 .
##
## MCMC diagnostics shown here are from the last round of simulation, prior to computation of final parameter estimates. Because the final estimates are refinements of those used for this simulation run, these diagnostics may understate model performance. To directly assess the performance of the final model on in-model statistics, please use the GOF command: gof(ergmFitObject, GOF=~model).