Ordinales Modell

Modellübersicht

##  Family: cumulative 
##   Links: mu = probit; disc = log 
## Formula: outcome ~ 1 + (1 | nuts2) 
##          disc ~ 1 + (1 | nuts2)
##    Data: cee18 (Number of observations: 6102) 
##   Draws: 3 chains, each with iter = 84000; warmup = 6000; thin = 6;
##          total post-warmup draws = 39000
## 
## Group-Level Effects: 
## ~nuts2 (Number of levels: 26) 
##                    Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## sd(Intercept)          0.96      0.31     0.49     1.68 1.00    21656    28430
## sd(disc_Intercept)     0.15      0.03     0.11     0.21 1.00    30556    34841
## 
## Population-Level Effects: 
##                Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept[1]      -4.51      1.28    -7.41    -2.44 1.00    20798    28218
## Intercept[2]      -3.59      1.02    -5.90    -1.93 1.00    20728    28174
## Intercept[3]      -2.56      0.74    -4.25    -1.36 1.00    20596    28599
## Intercept[4]      -1.64      0.50    -2.76    -0.85 1.00    20380    28598
## Intercept[5]      -0.74      0.28    -1.39    -0.30 1.00    19630    28906
## Intercept[6]       0.21      0.21    -0.16     0.66 1.00    18992    28311
## Intercept[7]       1.15      0.38     0.55     2.01 1.00    20553    28616
## Intercept[8]       2.32      0.68     1.23     3.88 1.00    20860    29267
## Intercept[9]       3.65      1.05     1.96     6.02 1.00    20957    28573
## Intercept[10]      5.03      1.43     2.70     8.27 1.00    20980    29103
## disc_Intercept    -0.51      0.28    -1.04     0.07 1.00    20973    28418
## 
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).

Trace-Plots

Trace-Plot

Autokorrelation-Plots

Autokorrelation-Plot

Posterior-Verteilung

Posterior-Plot

Ordinales Nullmodell

Modellübersicht

##  Family: cumulative 
##   Links: mu = probit; disc = log 
## Formula: outcome ~ 1 
##          disc ~ 1
##    Data: cee18 (Number of observations: 6102) 
##   Draws: 3 chains, each with iter = 84000; warmup = 6000; thin = 6;
##          total post-warmup draws = 39000
## 
## Population-Level Effects: 
##                Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept[1]      -4.03      1.15    -6.66    -2.15 1.00    13374    15460
## Intercept[2]      -3.32      0.95    -5.49    -1.77 1.00    13371    15690
## Intercept[3]      -2.47      0.71    -4.08    -1.32 1.00    13373    15838
## Intercept[4]      -1.66      0.48    -2.75    -0.89 1.00    13405    16223
## Intercept[5]      -0.87      0.25    -1.43    -0.46 1.00    13489    16087
## Intercept[6]       0.01      0.03    -0.05     0.07 1.00    38009    36265
## Intercept[7]       0.90      0.26     0.48     1.49 1.00    13443    16159
## Intercept[8]       2.05      0.59     1.10     3.39 1.00    13372    15892
## Intercept[9]       3.41      0.98     1.82     5.63 1.00    13359    15944
## Intercept[10]      4.81      1.38     2.56     7.98 1.00    13385    15623
## disc_Intercept    -0.56      0.29    -1.10     0.02 1.00    13346    15772
## 
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).

Trace-Plots

Trace-Plot

Autokorrelation-Plots

Autokorrelation-Plot

Posterior-Verteilung

Posterior-Plot

Metrisches Modell

Modellübersicht

##  Family: student 
##   Links: mu = identity; sigma = log; nu = identity 
## Formula: outcome ~ 1 + (1 | nuts2) 
##          sigma ~ 1 + (1 | nuts2)
##    Data: cee18 (Number of observations: 6102) 
##   Draws: 3 chains, each with iter = 66000; warmup = 6000; thin = 6;
##          total post-warmup draws = 30000
## 
## Group-Level Effects: 
## ~nuts2 (Number of levels: 26) 
##                     Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## sd(Intercept)           0.97      0.15     0.73     1.31 1.00    19414    24765
## sd(sigma_Intercept)     0.15      0.03     0.11     0.21 1.00    23862    26460
## 
## Population-Level Effects: 
##                 Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept           5.22      0.19     4.84     5.60 1.00    13332    21070
## sigma_Intercept     0.53      0.03     0.47     0.60 1.00    20680    25839
## 
## Family Specific Parameters: 
##    Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## nu    84.39     36.44    36.45   175.38 1.00    30316    29633
## 
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).

Trace-Plots

Trace-Plot

Autokorrelation-Plots

Autokorrelation-Plot

Posterior-Verteilung

Posterior-Plot

Metrisches Nullmodell

Modellübersicht

##  Family: student 
##   Links: mu = identity; sigma = log; nu = identity 
## Formula: outcome ~ 1 
##          sigma ~ 1
##    Data: cee18 (Number of observations: 6102) 
##   Draws: 3 chains, each with iter = 11500; warmup = 1500; thin = 1;
##          total post-warmup draws = 30000
## 
## Population-Level Effects: 
##                 Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept           5.35      0.03     5.30     5.40 1.00    19278    17569
## sigma_Intercept     0.67      0.01     0.65     0.69 1.00    20513    18800
## 
## Family Specific Parameters: 
##    Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## nu   133.80     46.53    64.95   244.03 1.00    18893    18156
## 
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).

Trace-Plots

Trace-Plot

Autokorrelation-Plots

Autokorrelation-Plot

Posterior-Verteilung

Posterior-Plot

Copyright:
Benedikt Philipp Kleer, 2022 (Online-Appendix zur Promotion, eingereicht am 14. September 2022, Fachbereich 03, Justus-Liebig-Universität Gießen).