strengejacke
commented
5 years ago Not sure, though, why pd() removes the .Intercept. suffix from the output of the random effects, while other methods don't?
Closed strengejacke closed 5 years ago
strengejacke
commented
5 years ago Not sure, though, why pd() removes the .Intercept. suffix from the output of the random effects, while other methods don't?
mattansb
commented
5 years ago @strengejacke you can give bf_params a go - the printing method currently is a bit messy, but I swear all the pieces are there for a reason!
One minor styling note - too many empty lines? So maybes eg this:
#> # MAP-based p-value
#>
#> # fixed effects, conditional component
#>
#> Parameter p_MAP
#> Intercept 0.212
#> child 0.000
#> camper 0.000
#>
#> # random effects, conditional component
#>
#> Parameter p_MAP
#> r_persons.1.Intercept. 0.2119
#> r_persons.2.Intercept. 0.6509
#> r_persons.3.Intercept. 0.9571
#> r_persons.4.Intercept. 0.0854can become this (especially considering as the headings are colored, right?):
#> # MAP-based p-value
#> # fixed effects, conditional component
#> Parameter p_MAP
#> Intercept 0.212
#> child 0.000
#> camper 0.000
#>
#> # random effects, conditional component
#> Parameter p_MAP
#> r_persons.1.Intercept. 0.2119
#> r_persons.2.Intercept. 0.6509
#> r_persons.3.Intercept. 0.9571
#> r_persons.4.Intercept. 0.0854Yes, I'm a bit undecided, it's really a matter of taste and I don't mind changing it. Essentially, all print-metods for multiple components/effects call print_data_frame(), where we can fix it once for all print-outs:
https://github.com/easystats/bayestestR/blob/master/R/utils_print_data_frame.R
The related code-line is here.
strengejacke
commented
5 years ago Fix
library(bayestestR)
library(insight)
m1 <- download_model("brms_zi_3")
#> Registered S3 method overwritten by 'xts':
#> method from
#> as.zoo.xts zoo
m2 <- download_model("stanreg_glm_2")
m3 <- download_model("stanreg_merMod_5")
# HDI, brms-model ---------------
hdi(m1)
#> # Highest Density Interval
#>
#> Parameter 89% HDI
#> Intercept [ 0.05, 2.27]
#> child [-1.32, -0.98]
#> camper [ 0.59, 0.86]
hdi(m1, effects = "all")
#> # Highest Density Interval
#>
#> # fixed effects, conditional component
#>
#> Parameter 89% HDI
#> Intercept [ 0.05, 2.27]
#> child [-1.32, -0.98]
#> camper [ 0.59, 0.86]
#>
#> # random effects, conditional component
#>
#> Parameter 89% HDI
#> persons 1 [-2.55, -0.03]
#> persons 2 [-1.45, 1.01]
#> persons 3 [-0.73, 1.59]
#> persons 4 [ 0.29, 2.54]
hdi(m1, effects = "all", component = "all")
#> # Highest Density Interval
#>
#> # fixed effects, conditional component
#>
#> Parameter 89% HDI
#> Intercept [ 0.05, 2.27]
#> child [-1.32, -0.98]
#> camper [ 0.59, 0.86]
#>
#> # fixed effects, zero-inflation component
#>
#> Parameter 89% HDI
#> Intercept [-1.89, 0.22]
#> child [ 1.30, 2.30]
#> camper [-1.34, -0.23]
#>
#> # random effects, conditional component
#>
#> Parameter 89% HDI
#> persons 1 [-2.55, -0.03]
#> persons 2 [-1.45, 1.01]
#> persons 3 [-0.73, 1.59]
#> persons 4 [ 0.29, 2.54]
#>
#> # random effects, zero-inflation component
#>
#> Parameter 89% HDI
#> persons 1 [ 0.37, 2.66]
#> persons 2 [-0.73, 1.49]
#> persons 3 [-1.16, 1.13]
#> persons 4 [-2.46, -0.06]
# HDI, rstanarm-model---------------
hdi(m2)
#> # Highest Density Interval
#>
#> Parameter 89% HDI
#> (Intercept) [ 5.24, 14.26]
#> wt [-0.89, 2.32]
#> cyl [-3.03, -0.91]
hdi(m2, effects = "all")
#> # Highest Density Interval
#>
#> Parameter 89% HDI
#> (Intercept) [ 5.24, 14.26]
#> wt [-0.89, 2.32]
#> cyl [-3.03, -0.91]
hdi(m2, effects = "all", component = "all")
#> # Highest Density Interval
#>
#> Parameter 89% HDI
#> (Intercept) [ 5.24, 14.26]
#> wt [-0.89, 2.32]
#> cyl [-3.03, -0.91]
# HDI, rstanarm-model---------------
hdi(m3)
#> # Highest Density Interval
#>
#> Parameter 89% HDI
#> (Intercept) [-2.45, -0.52]
#> size [-0.04, 0.05]
#> period2 [-1.50, -0.51]
#> period3 [-1.72, -0.63]
#> period4 [-2.27, -0.87]
hdi(m3, effects = "all")
#> # Highest Density Interval
#>
#> # fixed effects, conditional component
#>
#> Parameter 89% HDI
#> (Intercept) [-2.45, -0.52]
#> size [-0.04, 0.05]
#> period2 [-1.50, -0.51]
#> period3 [-1.72, -0.63]
#> period4 [-2.27, -0.87]
#>
#> # random effects, conditional component
#>
#> Parameter 89% HDI
#> herd:1 [-0.06, 1.35]
#> herd:2 [-1.09, 0.33]
#> herd:3 [-0.16, 1.01]
#> herd:4 [-0.80, 0.75]
#> herd:5 [-0.96, 0.36]
#> herd:6 [-1.20, 0.25]
#> herd:7 [ 0.18, 1.57]
#> herd:8 [-0.21, 1.38]
#> herd:9 [-1.21, 0.64]
#> herd:10 [-1.22, 0.19]
#> herd:11 [-0.76, 0.64]
#> herd:12 [-0.96, 0.69]
#> herd:13 [-1.48, -0.01]
#> herd:14 [ 0.17, 1.74]
#> herd:15 [-1.36, 0.20]
hdi(m3, effects = "all", component = "all")
#> # Highest Density Interval
#>
#> # fixed effects, conditional component
#>
#> Parameter 89% HDI
#> (Intercept) [-2.45, -0.52]
#> size [-0.04, 0.05]
#> period2 [-1.50, -0.51]
#> period3 [-1.72, -0.63]
#> period4 [-2.27, -0.87]
#>
#> # random effects, conditional component
#>
#> Parameter 89% HDI
#> herd:1 [-0.06, 1.35]
#> herd:2 [-1.09, 0.33]
#> herd:3 [-0.16, 1.01]
#> herd:4 [-0.80, 0.75]
#> herd:5 [-0.96, 0.36]
#> herd:6 [-1.20, 0.25]
#> herd:7 [ 0.18, 1.57]
#> herd:8 [-0.21, 1.38]
#> herd:9 [-1.21, 0.64]
#> herd:10 [-1.22, 0.19]
#> herd:11 [-0.76, 0.64]
#> herd:12 [-0.96, 0.69]
#> herd:13 [-1.48, -0.01]
#> herd:14 [ 0.17, 1.74]
#> herd:15 [-1.36, 0.20]
# pd, brms-model ---------------
pd(m1)
#> # Probability of Direction (pd)
#>
#> Parameter pd
#> Intercept 94.00%
#> child 100.00%
#> camper 100.00%
pd(m1, effects = "all")
#> # Probability of Direction (pd)
#>
#> # fixed effects, conditional component
#>
#> Parameter pd
#> Intercept 94.00%
#> child 100.00%
#> camper 100.00%
#>
#> # random effects, conditional component
#>
#> Parameter pd
#> persons 1 94.00%
#> persons 2 66.00%
#> persons 3 70.80%
#> persons 4 96.00%
pd(m1, effects = "all", component = "all")
#> # Probability of Direction (pd)
#>
#> # fixed effects, conditional component
#>
#> Parameter pd
#> Intercept 94.00%
#> child 100.00%
#> camper 100.00%
#>
#> # fixed effects, zero-inflation component
#>
#> Parameter pd
#> Intercept 87.60%
#> child 100.00%
#> camper 99.20%
#>
#> # random effects, conditional component
#>
#> Parameter pd
#> persons 1 94.00%
#> persons 2 66.00%
#> persons 3 70.80%
#> persons 4 96.00%
#>
#> # random effects, zero-inflation component
#>
#> Parameter pd
#> persons 1 95.60%
#> persons 2 72.40%
#> persons 3 58.00%
#> persons 4 97.20%
# pd, rstanarm-model---------------
pd(m2)
#> # Probability of Direction (pd)
#>
#> Parameter pd
#> (Intercept) 100.00%
#> wt 76.80%
#> cyl 100.00%
pd(m2, effects = "all")
#> # Probability of Direction (pd)
#>
#> Parameter pd
#> (Intercept) 100.00%
#> wt 76.80%
#> cyl 100.00%
pd(m2, effects = "all", component = "all")
#> # Probability of Direction (pd)
#>
#> Parameter pd
#> (Intercept) 100.00%
#> wt 76.80%
#> cyl 100.00%
# pd, rstanarm-model---------------
pd(m3)
#> # Probability of Direction (pd)
#>
#> Parameter pd
#> (Intercept) 99.40%
#> size 58.20%
#> period2 100.00%
#> period3 100.00%
#> period4 100.00%
pd(m3, effects = "all")
#> # Probability of Direction (pd)
#>
#> # fixed effects, conditional component
#>
#> Parameter pd
#> (Intercept) 99.40%
#> size 58.20%
#> period2 100.00%
#> period3 100.00%
#> period4 100.00%
#>
#> # random effects, conditional component
#>
#> Parameter pd
#> herd:1 92.40%
#> herd:2 81.40%
#> herd:3 88.40%
#> herd:4 52.60%
#> herd:5 73.20%
#> herd:6 84.80%
#> herd:7 99.20%
#> herd:8 85.00%
#> herd:9 63.80%
#> herd:10 93.60%
#> herd:11 61.80%
#> herd:12 52.80%
#> herd:13 97.40%
#> herd:14 98.60%
#> herd:15 88.60%
pd(m3, effects = "all", component = "all")
#> # Probability of Direction (pd)
#>
#> # fixed effects, conditional component
#>
#> Parameter pd
#> (Intercept) 99.40%
#> size 58.20%
#> period2 100.00%
#> period3 100.00%
#> period4 100.00%
#>
#> # random effects, conditional component
#>
#> Parameter pd
#> herd:1 92.40%
#> herd:2 81.40%
#> herd:3 88.40%
#> herd:4 52.60%
#> herd:5 73.20%
#> herd:6 84.80%
#> herd:7 99.20%
#> herd:8 85.00%
#> herd:9 63.80%
#> herd:10 93.60%
#> herd:11 61.80%
#> herd:12 52.80%
#> herd:13 97.40%
#> herd:14 98.60%
#> herd:15 88.60%
# point_estimate, brms-model ---------------
point_estimate(m1)
#> # Point Estimates
#>
#> Parameter Median
#> Intercept 1.32
#> child -1.16
#> camper 0.73
point_estimate(m1, centrality = "all", effects = "all")
#> # Point Estimates
#>
#> # fixed effects, conditional component
#>
#> Parameter Median Mean MAP
#> Intercept 1.32 1.19 1.45
#> child -1.16 -1.16 -1.18
#> camper 0.73 0.73 0.74
#>
#> # random effects, conditional component
#>
#> Parameter Median Mean MAP
#> persons 1 -1.32 -1.23 -1.40
#> persons 2 -0.38 -0.26 -0.54
#> persons 3 0.31 0.44 0.14
#> persons 4 1.21 1.33 1.03
point_estimate(m1, centrality = "all", effects = "all", component = "all")
#> # Point Estimates
#>
#> # fixed effects, conditional component
#>
#> Parameter Median Mean MAP
#> Intercept 1.32 1.19 1.45
#> child -1.16 -1.16 -1.18
#> camper 0.73 0.73 0.74
#>
#> # fixed effects, zero-inflation component
#>
#> Parameter Median Mean MAP
#> Intercept -0.78 -0.73 -0.89
#> child 1.89 1.88 1.91
#> camper -0.84 -0.84 -0.78
#>
#> # random effects, conditional component
#>
#> Parameter Median Mean MAP
#> persons 1 -1.32 -1.23 -1.40
#> persons 2 -0.38 -0.26 -0.54
#> persons 3 0.31 0.44 0.14
#> persons 4 1.21 1.33 1.03
#>
#> # random effects, zero-inflation component
#>
#> Parameter Median Mean MAP
#> persons 1 1.35 1.32 1.37
#> persons 2 0.38 0.36 0.51
#> persons 3 -0.12 -0.14 -0.10
#> persons 4 -1.17 -1.27 -1.02
# point_estimate, rstanarm-model---------------
point_estimate(m2)
#> # Point Estimates
#>
#> Parameter Median
#> (Intercept) 9.33
#> wt 0.76
#> cyl -1.99
point_estimate(m2, centrality = "all", effects = "all")
#> # Point Estimates
#>
#> Parameter Median Mean MAP
#> (Intercept) 9.33 9.64 9.04
#> wt 0.76 0.71 0.97
#> cyl -1.99 -2.02 -1.98
point_estimate(m2, centrality = "all", effects = "all", component = "all")
#> # Point Estimates
#>
#> Parameter Median Mean MAP
#> (Intercept) 9.33 9.64 9.04
#> wt 0.76 0.71 0.97
#> cyl -1.99 -2.02 -1.98
# point_estimate, rstanarm-model---------------
point_estimate(m3)
#> # Point Estimates
#>
#> Parameter Median
#> (Intercept) -1.4979
#> size 0.0045
#> period2 -0.9709
#> period3 -1.1121
#> period4 -1.6221
point_estimate(m3, centrality = "all", effects = "all")
#> # Point Estimates
#>
#> # fixed effects, conditional component
#>
#> Parameter Median Mean MAP
#> (Intercept) -1.4979 -1.5332 -1.2845
#> size 0.0045 0.0064 0.0035
#> period2 -0.9709 -0.9869 -0.9355
#> period3 -1.1121 -1.1209 -1.1829
#> period4 -1.6221 -1.6192 -1.6228
#>
#> # random effects, conditional component
#>
#> Parameter Median Mean MAP
#> herd:1 0.586 0.604 0.582
#> herd:2 -0.374 -0.400 -0.182
#> herd:3 0.376 0.400 0.308
#> herd:4 0.036 0.021 0.078
#> herd:5 -0.263 -0.287 -0.110
#> herd:6 -0.439 -0.471 -0.358
#> herd:7 0.887 0.920 0.861
#> herd:8 0.523 0.515 0.530
#> herd:9 -0.184 -0.246 -0.055
#> herd:10 -0.631 -0.639 -0.562
#> herd:11 -0.134 -0.137 -0.154
#> herd:12 -0.051 -0.085 0.108
#> herd:13 -0.786 -0.806 -0.742
#> herd:14 0.973 1.019 0.871
#> herd:15 -0.560 -0.601 -0.472
point_estimate(m3, centrality = "all", effects = "all", component = "all")
#> # Point Estimates
#>
#> # fixed effects, conditional component
#>
#> Parameter Median Mean MAP
#> (Intercept) -1.4979 -1.5332 -1.2845
#> size 0.0045 0.0064 0.0035
#> period2 -0.9709 -0.9869 -0.9355
#> period3 -1.1121 -1.1209 -1.1829
#> period4 -1.6221 -1.6192 -1.6228
#>
#> # random effects, conditional component
#>
#> Parameter Median Mean MAP
#> herd:1 0.586 0.604 0.582
#> herd:2 -0.374 -0.400 -0.182
#> herd:3 0.376 0.400 0.308
#> herd:4 0.036 0.021 0.078
#> herd:5 -0.263 -0.287 -0.110
#> herd:6 -0.439 -0.471 -0.358
#> herd:7 0.887 0.920 0.861
#> herd:8 0.523 0.515 0.530
#> herd:9 -0.184 -0.246 -0.055
#> herd:10 -0.631 -0.639 -0.562
#> herd:11 -0.134 -0.137 -0.154
#> herd:12 -0.051 -0.085 0.108
#> herd:13 -0.786 -0.806 -0.742
#> herd:14 0.973 1.019 0.871
#> herd:15 -0.560 -0.601 -0.472
# p_map, brms-model ---------------
p_map(m1)
#> # MAP-based p-value
#>
#> Parameter p_MAP
#> Intercept 0.212
#> child 0.000
#> camper 0.000
p_map(m1, effects = "all")
#> # MAP-based p-value
#>
#> # fixed effects, conditional component
#>
#> Parameter p_MAP
#> Intercept 0.212
#> child 0.000
#> camper 0.000
#>
#> # random effects, conditional component
#>
#> Parameter p_MAP
#> persons 1 0.2119
#> persons 2 0.6509
#> persons 3 0.9571
#> persons 4 0.0854
p_map(m1, effects = "all", component = "all")
#> # MAP-based p-value
#>
#> # fixed effects, conditional component
#>
#> Parameter p_MAP
#> Intercept 0.212
#> child 0.000
#> camper 0.000
#>
#> # fixed effects, zero-inflation component
#>
#> Parameter p_MAP
#> Intercept 0.4869
#> child 0.0000
#> camper 0.0574
#>
#> # random effects, conditional component
#>
#> Parameter p_MAP
#> persons 1 0.2119
#> persons 2 0.6509
#> persons 3 0.9571
#> persons 4 0.0854
#>
#> # random effects, zero-inflation component
#>
#> Parameter p_MAP
#> persons 1 0.181
#> persons 2 0.802
#> persons 3 0.984
#> persons 4 0.166
# p_map, rstanarm-model---------------
p_map(m2)
#> # MAP-based p-value
#>
#> Parameter p_MAP
#> (Intercept) 0.000
#> wt 0.698
#> cyl 0.000
p_map(m2, effects = "all")
#> # MAP-based p-value
#>
#> Parameter p_MAP
#> (Intercept) 0.000
#> wt 0.698
#> cyl 0.000
p_map(m2, effects = "all", component = "all")
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> # MAP-based p-value
#>
#> Parameter p_MAP
#> (Intercept) 0.000
#> wt 0.698
#> cyl 0.000
# p_map, rstanarm-model---------------
p_map(m3)
#> # MAP-based p-value
#>
#> Parameter p_MAP
#> (Intercept) 0.0476
#> size 0.9907
#> period2 0.0000
#> period3 0.0000
#> period4 0.0000
p_map(m3, effects = "all")
#> # MAP-based p-value
#>
#> # fixed effects, conditional component
#>
#> Parameter p_MAP
#> (Intercept) 0.0476
#> size 0.9907
#> period2 0.0000
#> period3 0.0000
#> period4 0.0000
#>
#> # random effects, conditional component
#>
#> Parameter p_MAP
#> herd:1 0.424
#> herd:2 0.818
#> herd:3 0.507
#> herd:4 0.990
#> herd:5 0.949
#> herd:6 0.619
#> herd:7 0.104
#> herd:8 0.572
#> herd:9 0.994
#> herd:10 0.449
#> herd:11 0.950
#> herd:12 0.956
#> herd:13 0.284
#> herd:14 0.107
#> herd:15 0.588
p_map(m3, effects = "all", component = "all")
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> Warning: In density.default(x, n = precision, bw = bw, from = x_range[1],
#> to = x_range[2], ...) :
#> extra argument 'component' will be disregarded
#> # MAP-based p-value
#>
#> # fixed effects, conditional component
#>
#> Parameter p_MAP
#> (Intercept) 0.0476
#> size 0.9907
#> period2 0.0000
#> period3 0.0000
#> period4 0.0000
#>
#> # random effects, conditional component
#>
#> Parameter p_MAP
#> herd:1 0.424
#> herd:2 0.818
#> herd:3 0.507
#> herd:4 0.990
#> herd:5 0.949
#> herd:6 0.619
#> herd:7 0.104
#> herd:8 0.572
#> herd:9 0.994
#> herd:10 0.449
#> herd:11 0.950
#> herd:12 0.956
#> herd:13 0.284
#> herd:14 0.107
#> herd:15 0.588
# describe_posterior, brms-model ---------------
describe_posterior(m1)
#> # Description of Posterior Distributions
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> Intercept 1.319 89 0.0495 2.275 0.94 89 -0.1 0.1
#> child -1.162 89 -1.3201 -0.980 1.00 89 -0.1 0.1
#> camper 0.727 89 0.5875 0.858 1.00 89 -0.1 0.1
#> ROPE_Percentage ESS Rhat
#> 0.0223 78 1.005
#> 0.0000 172 0.996
#> 0.0000 233 0.996
describe_posterior(m1, effects = "all")
#> # Description of Posterior Distributions
#>
#> # fixed effects, conditional component
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> Intercept 1.319 89 0.0495 2.275 0.94 89 -0.1 0.1
#> child -1.162 89 -1.3201 -0.980 1.00 89 -0.1 0.1
#> camper 0.727 89 0.5875 0.858 1.00 89 -0.1 0.1
#> ROPE_Percentage ESS Rhat
#> 0.0223 78 1.005
#> 0.0000 172 0.996
#> 0.0000 233 0.996
#>
#> # random effects, conditional component
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> persons 1 -1.315 89 -2.555 -0.0314 0.940 89 -0.1 0.1
#> persons 2 -0.380 89 -1.451 1.0085 0.660 89 -0.1 0.1
#> persons 3 0.307 89 -0.728 1.5882 0.708 89 -0.1 0.1
#> persons 4 1.207 89 0.290 2.5374 0.960 89 -0.1 0.1
#> ROPE_Percentage ESS Rhat
#> 0.0179 80 1.00
#> 0.0848 78 1.01
#> 0.1205 77 1.00
#> 0.0000 78 1.00
describe_posterior(m1, effects = "all", component = "all")
#> # Description of Posterior Distributions
#>
#> # fixed effects, conditional component
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> Intercept 1.319 89 0.0495 2.275 0.94 89 -0.1 0.1
#> child -1.162 89 -1.3201 -0.980 1.00 89 -0.1 0.1
#> camper 0.727 89 0.5875 0.858 1.00 89 -0.1 0.1
#> ROPE_Percentage ESS Rhat
#> 0.0223 78 1.005
#> 0.0000 172 0.996
#> 0.0000 233 0.996
#>
#> # fixed effects, zero-inflation component
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> Intercept -0.778 89 -1.89 0.218 0.876 89 -0.1 0.1
#> child 1.888 89 1.30 2.304 1.000 89 -0.1 0.1
#> camper -0.840 89 -1.34 -0.231 0.992 89 -0.1 0.1
#> ROPE_Percentage ESS Rhat
#> 0.0625 92 1.004
#> 0.0000 72 1.015
#> 0.0000 182 0.998
#>
#> # random effects, conditional component
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> persons 1 -1.315 89 -2.555 -0.0314 0.940 89 -0.1 0.1
#> persons 2 -0.380 89 -1.451 1.0085 0.660 89 -0.1 0.1
#> persons 3 0.307 89 -0.728 1.5882 0.708 89 -0.1 0.1
#> persons 4 1.207 89 0.290 2.5374 0.960 89 -0.1 0.1
#> ROPE_Percentage ESS Rhat
#> 0.0179 80 1.00
#> 0.0848 78 1.01
#> 0.1205 77 1.00
#> 0.0000 78 1.00
#>
#> # random effects, zero-inflation component
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> persons 1 1.355 89 0.368 2.6592 0.956 89 -0.1 0.1
#> persons 2 0.382 89 -0.726 1.4880 0.724 89 -0.1 0.1
#> persons 3 -0.117 89 -1.162 1.1283 0.580 89 -0.1 0.1
#> persons 4 -1.166 89 -2.462 -0.0609 0.972 89 -0.1 0.1
#> ROPE_Percentage ESS Rhat
#> 0.0000 91 1.005
#> 0.1205 99 1.000
#> 0.1429 94 0.997
#> 0.0134 113 0.997
# describe_posterior, rstanarm-model---------------
describe_posterior(m2)
#> Possible multicollinearity between cyl and wt (r = 0.7). This might lead to inappropriate results. See 'Details' in '?rope'.
#> # Description of Posterior Distributions
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> (Intercept) 9.333 89 5.243 14.263 1.000 89 -0.181 0.181
#> wt 0.759 89 -0.891 2.319 0.768 89 -0.181 0.181
#> cyl -1.985 89 -3.032 -0.907 1.000 89 -0.181 0.181
#> ROPE_Percentage ESS Rhat Prior_Distribution Prior_Location Prior_Scale
#> 0.000 2678 1 normal 0 10.00
#> 0.111 1745 1 normal 0 2.56
#> 0.000 1871 1 normal 0 1.40
describe_posterior(m2, effects = "all")
#> Possible multicollinearity between cyl and wt (r = 0.7). This might lead to inappropriate results. See 'Details' in '?rope'.
#> # Description of Posterior Distributions
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> (Intercept) 9.333 89 5.243 14.263 1.000 89 -0.181 0.181
#> wt 0.759 89 -0.891 2.319 0.768 89 -0.181 0.181
#> cyl -1.985 89 -3.032 -0.907 1.000 89 -0.181 0.181
#> ROPE_Percentage ESS Rhat Prior_Distribution Prior_Location Prior_Scale
#> 0.000 2678 1 normal 0 10.00
#> 0.111 1745 1 normal 0 2.56
#> 0.000 1871 1 normal 0 1.40
describe_posterior(m2, effects = "all", component = "all")
#> Possible multicollinearity between cyl and wt (r = 0.7). This might lead to inappropriate results. See 'Details' in '?rope'.
#> # Description of Posterior Distributions
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> (Intercept) 9.333 89 5.243 14.263 1.000 89 -0.181 0.181
#> wt 0.759 89 -0.891 2.319 0.768 89 -0.181 0.181
#> cyl -1.985 89 -3.032 -0.907 1.000 89 -0.181 0.181
#> ROPE_Percentage ESS Rhat Prior_Distribution Prior_Location Prior_Scale
#> 0.000 2678 1 normal 0 10.00
#> 0.111 1745 1 normal 0 2.56
#> 0.000 1871 1 normal 0 1.40
# describe_posterior, rstanarm-model---------------
describe_posterior(m3)
#> # Description of Posterior Distributions
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> (Intercept) -1.49792 89 -2.4521 -0.5200 0.994 89 -0.181 0.181
#> size 0.00452 89 -0.0409 0.0478 0.582 89 -0.181 0.181
#> period2 -0.97091 89 -1.4971 -0.5142 1.000 89 -0.181 0.181
#> period3 -1.11212 89 -1.7238 -0.6297 1.000 89 -0.181 0.181
#> period4 -1.62208 89 -2.2674 -0.8655 1.000 89 -0.181 0.181
#> ROPE_Percentage ESS Rhat Prior_Distribution Prior_Location Prior_Scale
#> 0 201 1.008 normal 0 10.0
#> 1 282 1.004 normal 0 2.5
#> 0 537 0.999 normal 0 2.5
#> 0 648 1.000 normal 0 2.5
#> 0 537 0.997 normal 0 2.5
describe_posterior(m3, effects = "all")
#> # Description of Posterior Distributions
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> (Intercept) -1.49792 89 -2.4521 -0.5200 0.994 89 -0.181 0.181
#> size 0.00452 89 -0.0409 0.0478 0.582 89 -0.181 0.181
#> period2 -0.97091 89 -1.4971 -0.5142 1.000 89 -0.181 0.181
#> period3 -1.11212 89 -1.7238 -0.6297 1.000 89 -0.181 0.181
#> period4 -1.62208 89 -2.2674 -0.8655 1.000 89 -0.181 0.181
#> ROPE_Percentage ESS Rhat Prior_Distribution Prior_Location Prior_Scale
#> 0 201 1.008 normal 0 10.0
#> 1 282 1.004 normal 0 2.5
#> 0 537 0.999 normal 0 2.5
#> 0 648 1.000 normal 0 2.5
#> 0 537 0.997 normal 0 2.5
describe_posterior(m3, effects = "all", component = "all")
#> # Description of Posterior Distributions
#>
#> Parameter Median CI CI_low CI_high pd ROPE_CI ROPE_low ROPE_high
#> (Intercept) -1.49792 89 -2.4521 -0.5200 0.994 89 -0.181 0.181
#> size 0.00452 89 -0.0409 0.0478 0.582 89 -0.181 0.181
#> period2 -0.97091 89 -1.4971 -0.5142 1.000 89 -0.181 0.181
#> period3 -1.11212 89 -1.7238 -0.6297 1.000 89 -0.181 0.181
#> period4 -1.62208 89 -2.2674 -0.8655 1.000 89 -0.181 0.181
#> ROPE_Percentage ESS Rhat Prior_Distribution Prior_Location Prior_Scale
#> 0 201 1.008 normal 0 10.0
#> 1 282 1.004 normal 0 2.5
#> 0 537 0.999 normal 0 2.5
#> 0 648 1.000 normal 0 2.5
#> 0 537 0.997 normal 0 2.5
# bayesfactor_parameters, brms-model ---------------
bayesfactor_parameters(m1, m1)
#> Loading required namespace: logspline
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> b_Intercept 1 fixed conditional
#> b_child 1 fixed conditional
#> b_camper 1 fixed conditional
#>
#> * Evidence Against The Null: [0]
bayesfactor_parameters(m1, m1, effects = "all")
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> b_Intercept 1 fixed conditional
#> b_child 1 fixed conditional
#> b_camper 1 fixed conditional
#> r_persons.1.Intercept. 1 random conditional
#> r_persons.2.Intercept. 1 random conditional
#> r_persons.3.Intercept. 1 random conditional
#> r_persons.4.Intercept. 1 random conditional
#>
#> * Evidence Against The Null: [0]
bayesfactor_parameters(m1, m1, effects = "all", component = "all")
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> b_Intercept 1 fixed conditional
#> b_child 1 fixed conditional
#> b_camper 1 fixed conditional
#> r_persons.1.Intercept. 1 random conditional
#> r_persons.2.Intercept. 1 random conditional
#> r_persons.3.Intercept. 1 random conditional
#> r_persons.4.Intercept. 1 random conditional
#> b_zi_Intercept 1 fixed zero_inflated
#> b_zi_child 1 fixed zero_inflated
#> b_zi_camper 1 fixed zero_inflated
#> r_persons__zi.1.Intercept. 1 random zero_inflated
#> r_persons__zi.2.Intercept. 1 random zero_inflated
#> r_persons__zi.3.Intercept. 1 random zero_inflated
#> r_persons__zi.4.Intercept. 1 random zero_inflated
#>
#> * Evidence Against The Null: [0]
# bayesfactor_parameters, rstanarm-model---------------
bayesfactor_parameters(m2, m2)
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> (Intercept) 1 fixed conditional
#> wt 1 fixed conditional
#> cyl 1 fixed conditional
#>
#> * Evidence Against The Null: [0]
bayesfactor_parameters(m2, m2, effects = "all")
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> (Intercept) 1 fixed conditional
#> wt 1 fixed conditional
#> cyl 1 fixed conditional
#>
#> * Evidence Against The Null: [0]
bayesfactor_parameters(m2, m2, effects = "all", component = "all")
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> (Intercept) 1 fixed conditional
#> wt 1 fixed conditional
#> cyl 1 fixed conditional
#>
#> * Evidence Against The Null: [0]
# bayesfactor_parameters, rstanarm-model---------------
bayesfactor_parameters(m3, m3)
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> (Intercept) 1 fixed conditional
#> size 1 fixed conditional
#> period2 1 fixed conditional
#> period3 1 fixed conditional
#> period4 1 fixed conditional
#>
#> * Evidence Against The Null: [0]
bayesfactor_parameters(m3, m3, effects = "all")
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> (Intercept) 1 fixed conditional
#> size 1 fixed conditional
#> period2 1 fixed conditional
#> period3 1 fixed conditional
#> period4 1 fixed conditional
#> b[(Intercept) herd:1] 1 random conditional
#> b[(Intercept) herd:2] 1 random conditional
#> b[(Intercept) herd:3] 1 random conditional
#> b[(Intercept) herd:4] 1 random conditional
#> b[(Intercept) herd:5] 1 random conditional
#> b[(Intercept) herd:6] 1 random conditional
#> b[(Intercept) herd:7] 1 random conditional
#> b[(Intercept) herd:8] 1 random conditional
#> b[(Intercept) herd:9] 1 random conditional
#> b[(Intercept) herd:10] 1 random conditional
#> b[(Intercept) herd:11] 1 random conditional
#> b[(Intercept) herd:12] 1 random conditional
#> b[(Intercept) herd:13] 1 random conditional
#> b[(Intercept) herd:14] 1 random conditional
#> b[(Intercept) herd:15] 1 random conditional
#>
#> * Evidence Against The Null: [0]
bayesfactor_parameters(m3, m3, effects = "all", component = "all")
#> # Bayes Factor (Savage-Dickey density ratio)
#>
#> Parameter Bayes Factor Effects Component
#> (Intercept) 1 fixed conditional
#> size 1 fixed conditional
#> period2 1 fixed conditional
#> period3 1 fixed conditional
#> period4 1 fixed conditional
#> b[(Intercept) herd:1] 1 random conditional
#> b[(Intercept) herd:2] 1 random conditional
#> b[(Intercept) herd:3] 1 random conditional
#> b[(Intercept) herd:4] 1 random conditional
#> b[(Intercept) herd:5] 1 random conditional
#> b[(Intercept) herd:6] 1 random conditional
#> b[(Intercept) herd:7] 1 random conditional
#> b[(Intercept) herd:8] 1 random conditional
#> b[(Intercept) herd:9] 1 random conditional
#> b[(Intercept) herd:10] 1 random conditional
#> b[(Intercept) herd:11] 1 random conditional
#> b[(Intercept) herd:12] 1 random conditional
#> b[(Intercept) herd:13] 1 random conditional
#> b[(Intercept) herd:14] 1 random conditional
#> b[(Intercept) herd:15] 1 random conditional
#>
#> * Evidence Against The Null: [0]Created on 2019-09-11 by the reprex package (v0.3.0)
This is a long example, so I though just for demonstration purposes, I show the print-out in this issue (instead of other issues, where this would clutter up the thread).
We could also fix the print for BF, what do you think? @mattansb @DominiqueMakowski
Created on 2019-09-11 by the reprex package (v0.3.0)