## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 297290 15.9 407500 21.8 350000 18.7
## Vcells 520128 4.0 905753 7.0 786426 6.0## Loading required package: bbmle
## Loading required package: stats4
## Loading required package: survival
## Loading required package: caret
## Loading required package: lattice## Warning: package 'lattice' was built under R version 3.1.3## Loading required package: ggplot2
##
## Attaching package: 'caret'
##
## The following object is masked from 'package:survival':
##
## cluster# RUM General Results
RUM_all <- run_RUM_model(2, 1, p=1)## [1] "Part 1 Opt-out.csv"
## initial value 5662.208659
## iter 10 value 2146.600613
## iter 20 value 944.734676
## final value 944.584835
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00296 0.00017 -17.061 0.00000
## eff_3x:B 1.60184 0.18115 8.842 0.00001
## eff_2x:B 1.75546 0.18089 9.705 0.00000
## eff_1x:B 2.07542 0.18862 11.003 0.00000
## eff_1bix:B 2.10136 0.18763 11.200 0.00000
## dosead1:B 0.16243 0.10863 1.495 0.16906
## sides1:B 0.38420 0.11231 3.421 0.00762
## type1:B -0.10512 0.10967 -0.958 0.36287
## beta.optout:B -5.41030 0.10625 -50.923 0.00000
##
## Log-Likelihood: -944.5848
## Iterations: 114.0000# RRM General Results
RRM_all <- RRM_boot_opt_out_generic(data.file = "Part 1 Opt-out.csv", verbose=TRUE,
seed_nbr = 1, p = 1, split_by_variable=FALSE)## initial value 5662.208659
## iter 10 value 2146.600613
## iter 20 value 944.734676
## final value 944.584835
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00296 0.00017 -17.061 0.00000
## eff_3x:B 1.60184 0.18115 8.842 0.00001
## eff_2x:B 1.75546 0.18089 9.705 0.00000
## eff_1x:B 2.07542 0.18862 11.003 0.00000
## eff_1bix:B 2.10136 0.18763 11.200 0.00000
## dosead1:B 0.16243 0.10863 1.495 0.16906
## sides1:B 0.38420 0.11231 3.421 0.00762
## type1:B -0.10512 0.10967 -0.958 0.36287
## beta.optout:B -5.41030 0.10625 -50.923 0.00000
##
## Log-Likelihood: -944.5848
## Iterations: 114.0000## Loading required package: data.table## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 1556553 83.2 2403845 128.4 2401145 128.3
## Vcells 2141456 16.4 3869736 29.6 3869736 29.6Optim summary of RUM - By Gender
RUM_op_wtp_gender_opt_out <- lapply(variable_levels, function(lv){
print(sprintf('Variable: %s Level: %s', variable, lv))
data.file = 'Part 1 Opt-out.csv'
run_RUM_model_generic(2, seed_nbr=1, variable, variable_level=lv, p=1)
})## [1] "Variable: gender Level: 1"
## initial value 2671.525027
## iter 10 value 611.824183
## iter 20 value 423.837886
## iter 20 value 423.837886
## final value 423.837886
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00316 0.00026 -12.11 0.00000
## eff_3x:B 1.24982 0.26502 4.72 0.00110
## eff_2x:B 1.38149 0.26480 5.22 0.00055
## eff_1x:B 1.80905 0.27400 6.60 0.00010
## eff_1bix:B 2.06883 0.28609 7.23 0.00005
## dosead1:B 0.17647 0.15966 1.11 0.29771
## sides1:B 0.53680 0.16715 3.21 0.01063
## type1:B -0.30330 0.16341 -1.86 0.09641
## beta.optout:B -5.58455 0.16740 -33.36 0.00000
##
## Log-Likelihood: -423.8379
## Iterations: 74.0000
## [1] "Variable: gender Level: 2"
## initial value 2986.195163
## iter 10 value 770.105044
## iter 20 value 513.299616
## final value 513.266728
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00294 0.00024 -12.268 0.00000
## eff_3x:B 1.92951 0.25302 7.626 0.00003
## eff_2x:B 2.14140 0.25588 8.369 0.00002
## eff_1x:B 2.35173 0.26388 8.912 0.00001
## eff_1bix:B 2.18701 0.25473 8.586 0.00001
## dosead1:B 0.14558 0.15092 0.965 0.35995
## sides1:B 0.22520 0.15541 1.449 0.18124
## type1:B 0.06424 0.15133 0.425 0.68116
## beta.optout:B -5.27593 0.13733 -38.417 0.00000
##
## Log-Likelihood: -513.2667
## Iterations: 109.0000names(RUM_op_wtp_gender_opt_out) <- variable_levels
save(RUM_op_wtp_gender_opt_out, file='RUM_op_wtp_gender_opt_out.RData')Optim Summary of RRM - By Gender
### Income RRM Model
source('fun_avg_wtp.R')
source('RRM_boot_opt_out_generic.R')
RRM_op_wtp_gender_opt_out <- lapply(variable_levels, function(inc){
print(sprintf('Variable: %s Level: %s', variable, inc))
data.file = 'Part 1 Opt-out.csv'
RRM_boot_opt_out_generic(data.file, seed_nbr=1, variable,
variable_level=inc, p=1,
fix_ethnicity = FALSE, split_by_variable = TRUE,
verbose=TRUE)
})## [1] "Variable: gender Level: 1"
## initial value 2671.525027
## iter 10 value 611.824183
## iter 20 value 423.837886
## iter 20 value 423.837886
## final value 423.837886
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00316 0.00026 -12.11 0.00000
## eff_3x:B 1.24982 0.26502 4.72 0.00110
## eff_2x:B 1.38149 0.26480 5.22 0.00055
## eff_1x:B 1.80905 0.27400 6.60 0.00010
## eff_1bix:B 2.06883 0.28609 7.23 0.00005
## dosead1:B 0.17647 0.15966 1.11 0.29771
## sides1:B 0.53680 0.16715 3.21 0.01063
## type1:B -0.30330 0.16341 -1.86 0.09641
## beta.optout:B -5.58455 0.16740 -33.36 0.00000
##
## Log-Likelihood: -423.8379
## Iterations: 74.0000
## [1] "Variable: gender Level: 2"
## initial value 2986.195163
## iter 10 value 770.105044
## iter 20 value 513.299616
## final value 513.266728
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00294 0.00024 -12.268 0.00000
## eff_3x:B 1.92951 0.25302 7.626 0.00003
## eff_2x:B 2.14140 0.25588 8.369 0.00002
## eff_1x:B 2.35173 0.26388 8.912 0.00001
## eff_1bix:B 2.18701 0.25473 8.586 0.00001
## dosead1:B 0.14558 0.15092 0.965 0.35995
## sides1:B 0.22520 0.15541 1.449 0.18124
## type1:B 0.06424 0.15133 0.425 0.68116
## beta.optout:B -5.27593 0.13733 -38.417 0.00000
##
## Log-Likelihood: -513.2667
## Iterations: 109.0000names(RRM_op_wtp_gender_opt_out) <- variable_levels
save(RRM_op_wtp_gender_opt_out, file='RRM_op_wtp_gender_opt_out.RData')Optim summary of RUM - By Income
RUM_op_wtp_income_opt_out <- lapply(variable_levels, function(lv){
print(sprintf('Variable: %s Level: %s', variable, lv))
data.file = 'Part 1 Opt-out.csv'
run_RUM_model_generic(2, seed_nbr=1, variable, variable_level=lv, p=1)
})## [1] "Variable: income Level: 1"
## initial value 1765.081807
## iter 10 value 402.028784
## iter 20 value 352.339813
## final value 352.305537
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00251 0.00028 -8.977 0.00001
## eff_3x:B 1.48023 0.29863 4.957 0.00078
## eff_2x:B 1.48531 0.30302 4.902 0.00085
## eff_1x:B 1.72063 0.30898 5.569 0.00035
## eff_1bix:B 1.59524 0.29944 5.327 0.00048
## dosead1:B 0.19071 0.18543 1.028 0.33057
## sides1:B 0.20055 0.19502 1.028 0.33063
## type1:B -0.11348 0.18351 -0.618 0.55164
## beta.optout:B -4.90413 0.15304 -32.046 0.00000
##
## Log-Likelihood: -352.3055
## Iterations: 133.0000
## [1] "Variable: income Level: 2"
## initial value 1504.318752
## iter 10 value 269.449081
## final value 257.008838
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00345 0.00037 -9.306 0.00001
## eff_3x:B 1.27162 0.34949 3.639 0.00541
## eff_2x:B 1.48004 0.34505 4.289 0.00202
## eff_1x:B 1.82635 0.37019 4.934 0.00081
## eff_1bix:B 1.51251 0.35048 4.316 0.00195
## dosead1:B 0.05297 0.21272 0.249 0.80895
## sides1:B 0.42218 0.21092 2.002 0.07635
## type1:B -0.09924 0.21906 -0.453 0.66126
## beta.optout:B -5.25694 0.19354 -27.162 0.00000
##
## Log-Likelihood: -257.0088
## Iterations: 79.0000
## [1] "Variable: income Level: 3"
## initial value 1073.016643
## iter 10 value 365.293038
## iter 20 value 154.890450
## iter 30 value 153.353076
## final value 153.346795
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00287 0.00039 -7.290 0.00005
## eff_3x:B 2.24800 0.47642 4.719 0.00109
## eff_2x:B 2.56906 0.49396 5.201 0.00056
## eff_1x:B 2.76525 0.49822 5.550 0.00036
## eff_1bix:B 3.25294 0.50768 6.407 0.00012
## dosead1:B 0.44489 0.27509 1.617 0.14028
## sides1:B 0.55307 0.28590 1.934 0.08506
## type1:B -0.18426 0.26843 -0.686 0.50973
## beta.optout:B -5.96138 0.30566 -19.503 0.00000
##
## Log-Likelihood: -153.3468
## Iterations: 113.0000
## [1] "Variable: income Level: 4"
## initial value 1306.326052
## iter 10 value 382.044054
## iter 20 value 149.984993
## final value 149.923603
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00406 0.00048 -8.4636 0.00001
## eff_3x:B 1.88172 0.44687 4.2108 0.00227
## eff_2x:B 2.39272 0.44223 5.4105 0.00043
## eff_1x:B 2.74667 0.47941 5.7293 0.00028
## eff_1bix:B 3.28593 0.51116 6.4284 0.00012
## dosead1:B 0.02067 0.25441 0.0813 0.93702
## sides1:B 0.68039 0.27819 2.4457 0.03701
## type1:B -0.18683 0.25314 -0.7380 0.47929
## beta.optout:B -6.56612 0.36678 -17.9020 0.00000
##
## Log-Likelihood: -149.9236
## Iterations: 108.0000names(RUM_op_wtp_income_opt_out) <- variable_levels
save(RUM_op_wtp_income_opt_out, file='RUM_op_wtp_income_opt_out.RData')Optim Summary of RRM - By Income
### Income RRM Model
source('fun_avg_wtp.R')
source('RRM_boot_opt_out_generic.R')
RRM_op_wtp_income_opt_out <- lapply(variable_levels, function(inc){
print(sprintf('Variable: %s Level: %s', variable, inc))
data.file = 'Part 1 Opt-out.csv'
RRM_boot_opt_out_generic(data.file, seed_nbr=1, variable,
variable_level=inc, p=1,
fix_ethnicity = FALSE, split_by_variable = TRUE,
verbose=TRUE)
})## [1] "Variable: income Level: 1"
## initial value 1765.081807
## iter 10 value 402.028784
## iter 20 value 352.339813
## final value 352.305537
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00251 0.00028 -8.977 0.00001
## eff_3x:B 1.48023 0.29863 4.957 0.00078
## eff_2x:B 1.48531 0.30302 4.902 0.00085
## eff_1x:B 1.72063 0.30898 5.569 0.00035
## eff_1bix:B 1.59524 0.29944 5.327 0.00048
## dosead1:B 0.19071 0.18543 1.028 0.33057
## sides1:B 0.20055 0.19502 1.028 0.33063
## type1:B -0.11348 0.18351 -0.618 0.55164
## beta.optout:B -4.90413 0.15304 -32.046 0.00000
##
## Log-Likelihood: -352.3055
## Iterations: 133.0000
## [1] "Variable: income Level: 2"
## initial value 1504.318752
## iter 10 value 269.449081
## final value 257.008838
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00345 0.00037 -9.306 0.00001
## eff_3x:B 1.27162 0.34949 3.639 0.00541
## eff_2x:B 1.48004 0.34505 4.289 0.00202
## eff_1x:B 1.82635 0.37019 4.934 0.00081
## eff_1bix:B 1.51251 0.35048 4.316 0.00195
## dosead1:B 0.05297 0.21272 0.249 0.80895
## sides1:B 0.42218 0.21092 2.002 0.07635
## type1:B -0.09924 0.21906 -0.453 0.66126
## beta.optout:B -5.25694 0.19354 -27.162 0.00000
##
## Log-Likelihood: -257.0088
## Iterations: 79.0000
## [1] "Variable: income Level: 3"
## initial value 1073.016643
## iter 10 value 365.293038
## iter 20 value 154.890450
## iter 30 value 153.353076
## final value 153.346795
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00287 0.00039 -7.290 0.00005
## eff_3x:B 2.24800 0.47642 4.719 0.00109
## eff_2x:B 2.56906 0.49396 5.201 0.00056
## eff_1x:B 2.76525 0.49822 5.550 0.00036
## eff_1bix:B 3.25294 0.50768 6.407 0.00012
## dosead1:B 0.44489 0.27509 1.617 0.14028
## sides1:B 0.55307 0.28590 1.934 0.08506
## type1:B -0.18426 0.26843 -0.686 0.50973
## beta.optout:B -5.96138 0.30566 -19.503 0.00000
##
## Log-Likelihood: -153.3468
## Iterations: 113.0000
## [1] "Variable: income Level: 4"
## initial value 1306.326052
## iter 10 value 382.044054
## iter 20 value 149.984993
## final value 149.923603
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00406 0.00048 -8.4636 0.00001
## eff_3x:B 1.88172 0.44687 4.2108 0.00227
## eff_2x:B 2.39272 0.44223 5.4105 0.00043
## eff_1x:B 2.74667 0.47941 5.7293 0.00028
## eff_1bix:B 3.28593 0.51116 6.4284 0.00012
## dosead1:B 0.02067 0.25441 0.0813 0.93702
## sides1:B 0.68039 0.27819 2.4457 0.03701
## type1:B -0.18683 0.25314 -0.7380 0.47929
## beta.optout:B -6.56612 0.36678 -17.9020 0.00000
##
## Log-Likelihood: -149.9236
## Iterations: 108.0000names(RRM_op_wtp_income_opt_out) <- variable_levels
save(RRM_op_wtp_income_opt_out, file='RRM_op_wtp_income_opt_out.RData')## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 1557195 83.2 2403845 128.4 2401145 128.3
## Vcells 2144407 16.4 3869736 29.6 3869736 29.6Optim summary of RUM - By Ethnicity
RUM_op_wtp_ethnic_opt_out <- lapply(variable_levels, function(lv){
print(sprintf('Variable: %s Level: %s', variable, lv))
data.file = 'Part 1 Opt-out.csv'
run_RUM_model_generic(2, seed_nbr=1, variable, variable_level=lv, p=1, fix_ethnicity=TRUE)
})## [1] "Variable: ethnic Level: 1"
##
## 1 2 3 4
## 3042 468 216 360
## initial value 4269.942483
## iter 10 value 1503.022145
## iter 20 value 658.107712
## final value 658.041469
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00322 0.00021 -15.244 0.00000
## eff_3x:B 1.69844 0.21381 7.944 0.00002
## eff_2x:B 1.75126 0.21536 8.132 0.00002
## eff_1x:B 2.24558 0.22796 9.851 0.00000
## eff_1bix:B 2.39500 0.22595 10.600 0.00000
## dosead1:B 0.23076 0.13085 1.764 0.11164
## sides1:B 0.34630 0.13268 2.610 0.02827
## type1:B 0.02763 0.12926 0.214 0.83551
## beta.optout:B -5.60841 0.13327 -42.084 0.00000
##
## Log-Likelihood: -658.0415
## Iterations: 119.0000
## [1] "Variable: ethnic Level: 2"
##
## 1 2 3 4
## 3042 468 216 360
## initial value 624.238037
## iter 10 value 152.747346
## iter 20 value 117.845730
## iter 20 value 117.845730
## final value 117.845730
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.0027 0.00051 -5.317 0.00048
## eff_3x:B 0.6973 0.51949 1.342 0.21241
## eff_2x:B 1.3969 0.54367 2.569 0.03022
## eff_1x:B 1.6155 0.48154 3.355 0.00846
## eff_1bix:B 1.9784 0.56674 3.491 0.00682
## dosead1:B 0.2889 0.30859 0.936 0.37366
## sides1:B 0.5549 0.35046 1.583 0.14780
## type1:B -1.0779 0.34971 -3.082 0.01309
## beta.optout:B -5.1495 0.28349 -18.165 0.00000
##
## Log-Likelihood: -117.8457
## Iterations: 70.0000
## [1] "Variable: ethnic Level: 3"
##
## 1 2 3 4
## 3042 468 216 360
## initial value 305.284022
## iter 10 value 105.761885
## iter 20 value 41.140444
## iter 30 value 40.870282
## iter 30 value 40.870282
## final value 40.866108
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00382 0.00089 -4.294 0.00201
## eff_3x:B 1.36234 1.04976 1.298 0.22664
## eff_2x:B 2.28405 0.78563 2.907 0.01739
## eff_1x:B 2.20589 1.01056 2.183 0.05691
## eff_1bix:B 1.63746 0.93426 1.753 0.11356
## dosead1:B -0.28208 0.53408 -0.528 0.61017
## sides1:B 1.15388 0.55816 2.067 0.06866
## type1:B -1.14243 0.57361 -1.992 0.07759
## beta.optout:B -6.25260 0.63275 -9.882 0.00000
##
## Log-Likelihood: -40.86611
## Iterations: 130.00000
## [1] "Variable: ethnic Level: 4"
##
## 1 2 3 4
## 3042 468 216 360
## initial value 453.744449
## iter 10 value 112.546675
## iter 20 value 96.749765
## final value 96.742084
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00282 0.0006 -4.716 0.00110
## eff_3x:B 2.24971 0.6206 3.625 0.00553
## eff_2x:B 2.50594 0.6643 3.772 0.00440
## eff_1x:B 2.20364 0.7511 2.934 0.01665
## eff_1bix:B 1.40622 0.5782 2.432 0.03784
## dosead1:B -0.24257 0.3779 -0.642 0.53694
## sides1:B 0.50339 0.3847 1.308 0.22315
## type1:B 0.48357 0.3891 1.243 0.24531
## beta.optout:B -4.63256 0.2664 -17.389 0.00000
##
## Log-Likelihood: -96.74208
## Iterations: 96.00000names(RUM_op_wtp_ethnic_opt_out) <- variable_levels
save(RUM_op_wtp_ethnic_opt_out, file='RUM_op_wtp_ethnic_opt_out.RData')Optim Summary of RRM - By Ethnicity
### Income RRM Model
source('fun_avg_wtp.R')
source('RRM_boot_opt_out_generic.R')
RRM_op_wtp_ethnic_opt_out <- lapply(variable_levels, function(inc){
print(sprintf('Variable: %s Level: %s', variable, inc))
data.file = 'Part 1 Opt-out.csv'
RRM_boot_opt_out_generic(data.file, seed_nbr=1, variable,
variable_level=inc, p=1,
fix_ethnicity = TRUE, split_by_variable = TRUE,
verbose=TRUE)
})## [1] "Variable: ethnic Level: 1"
##
## 1 2 3 4
## 3042 468 216 360
## initial value 4269.942483
## iter 10 value 1503.022145
## iter 20 value 658.107712
## final value 658.041469
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00322 0.00021 -15.244 0.00000
## eff_3x:B 1.69844 0.21381 7.944 0.00002
## eff_2x:B 1.75126 0.21536 8.132 0.00002
## eff_1x:B 2.24558 0.22796 9.851 0.00000
## eff_1bix:B 2.39500 0.22595 10.600 0.00000
## dosead1:B 0.23076 0.13085 1.764 0.11164
## sides1:B 0.34630 0.13268 2.610 0.02827
## type1:B 0.02763 0.12926 0.214 0.83551
## beta.optout:B -5.60841 0.13327 -42.084 0.00000
##
## Log-Likelihood: -658.0415
## Iterations: 119.0000
## [1] "Variable: ethnic Level: 2"
##
## 1 2 3 4
## 3042 468 216 360
## initial value 624.238037
## iter 10 value 152.747346
## iter 20 value 117.845730
## iter 20 value 117.845730
## final value 117.845730
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.0027 0.00051 -5.317 0.00048
## eff_3x:B 0.6973 0.51949 1.342 0.21241
## eff_2x:B 1.3969 0.54367 2.569 0.03022
## eff_1x:B 1.6155 0.48154 3.355 0.00846
## eff_1bix:B 1.9784 0.56674 3.491 0.00682
## dosead1:B 0.2889 0.30859 0.936 0.37366
## sides1:B 0.5549 0.35046 1.583 0.14780
## type1:B -1.0779 0.34971 -3.082 0.01309
## beta.optout:B -5.1495 0.28349 -18.165 0.00000
##
## Log-Likelihood: -117.8457
## Iterations: 70.0000
## [1] "Variable: ethnic Level: 3"
##
## 1 2 3 4
## 3042 468 216 360
## initial value 305.284022
## iter 10 value 105.761885
## iter 20 value 41.140444
## iter 30 value 40.870282
## iter 30 value 40.870282
## final value 40.866108
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00382 0.00089 -4.294 0.00201
## eff_3x:B 1.36234 1.04976 1.298 0.22664
## eff_2x:B 2.28405 0.78563 2.907 0.01739
## eff_1x:B 2.20589 1.01056 2.183 0.05691
## eff_1bix:B 1.63746 0.93426 1.753 0.11356
## dosead1:B -0.28208 0.53408 -0.528 0.61017
## sides1:B 1.15388 0.55816 2.067 0.06866
## type1:B -1.14243 0.57361 -1.992 0.07759
## beta.optout:B -6.25260 0.63275 -9.882 0.00000
##
## Log-Likelihood: -40.86611
## Iterations: 130.00000
## [1] "Variable: ethnic Level: 4"
##
## 1 2 3 4
## 3042 468 216 360
## initial value 453.744449
## iter 10 value 112.546675
## iter 20 value 96.749765
## final value 96.742084
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00282 0.0006 -4.716 0.00110
## eff_3x:B 2.24971 0.6206 3.625 0.00553
## eff_2x:B 2.50594 0.6643 3.772 0.00440
## eff_1x:B 2.20364 0.7511 2.934 0.01665
## eff_1bix:B 1.40622 0.5782 2.432 0.03784
## dosead1:B -0.24257 0.3779 -0.642 0.53694
## sides1:B 0.50339 0.3847 1.308 0.22315
## type1:B 0.48357 0.3891 1.243 0.24531
## beta.optout:B -4.63256 0.2664 -17.389 0.00000
##
## Log-Likelihood: -96.74208
## Iterations: 96.00000names(RRM_op_wtp_ethnic_opt_out) <- variable_levels
save(RRM_op_wtp_ethnic_opt_out, file='RRM_op_wtp_ethnic_opt_out.RData')## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 1557429 83.2 2403845 128.4 2403845 128.4
## Vcells 2145843 16.4 4143222 31.7 4143222 31.7Optim summary of RUM - By Age
RUM_op_wtp_age_opt_out <- lapply(variable_levels, function(lv){
print(sprintf('Variable: %s Level: %s', variable, lv))
data.file = 'Part 1 Opt-out.csv'
run_RUM_model_generic(2, seed_nbr=1, variable, variable_level=lv, p=1, fix_age=TRUE)
})## [1] "Variable: age_group Level: 1"
##
## 1 2 3
## 1044 1638 1404
## initial value 1396.709180
## iter 10 value 269.068602
## iter 20 value 244.075410
## final value 244.058570
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00348 0.00039 -8.889 0.00001
## eff_3x:B 2.14818 0.41044 5.234 0.00054
## eff_2x:B 2.58478 0.42878 6.028 0.00020
## eff_1x:B 2.18745 0.41618 5.256 0.00052
## eff_1bix:B 1.79972 0.41131 4.376 0.00178
## dosead1:B 0.06303 0.22321 0.282 0.78406
## sides1:B 0.32471 0.25841 1.257 0.24055
## type1:B 0.15647 0.23603 0.663 0.52398
## beta.optout:B -5.04396 0.18245 -27.646 0.00000
##
## Log-Likelihood: -244.0586
## Iterations: 111.0000
## [1] "Variable: age_group Level: 2"
##
## 1 2 3
## 1044 1638 1404
## initial value 2289.914317
## iter 10 value 639.890834
## final value 356.776574
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00336 0.0003 -11.25 0.00000
## eff_3x:B 1.30452 0.2792 4.67 0.00117
## eff_2x:B 1.46503 0.2813 5.21 0.00056
## eff_1x:B 2.02496 0.2992 6.77 0.00008
## eff_1bix:B 2.24768 0.2943 7.64 0.00003
## dosead1:B 0.42410 0.1837 2.31 0.04629
## sides1:B 0.43193 0.1810 2.39 0.04079
## type1:B -0.20918 0.1823 -1.15 0.28084
## beta.optout:B -5.56791 0.1784 -31.20 0.00000
##
## Log-Likelihood: -356.7766
## Iterations: 86.0000
## [1] "Variable: age_group Level: 3"
##
## 1 2 3
## 1044 1638 1404
## initial value 1966.608226
## iter 10 value 393.357769
## final value 325.947490
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00256 0.00027 -9.563 0.00001
## eff_3x:B 1.66045 0.30740 5.402 0.00043
## eff_2x:B 1.68981 0.29322 5.763 0.00027
## eff_1x:B 2.14615 0.31629 6.785 0.00008
## eff_1bix:B 2.18528 0.31808 6.870 0.00007
## dosead1:B 0.06238 0.18258 0.342 0.74046
## sides1:B 0.37734 0.18104 2.084 0.06680
## type1:B -0.12870 0.17948 -0.717 0.49152
## beta.optout:B -5.57859 0.19372 -28.797 0.00000
##
## Log-Likelihood: -325.9475
## Iterations: 87.0000names(RUM_op_wtp_age_opt_out) <- variable_levels
save(RUM_op_wtp_age_opt_out, file='RUM_op_wtp_age_opt_out.RData')Optim Summary of RRM - By Age
### Income RRM Model
source('fun_avg_wtp.R')
source('RRM_boot_opt_out_generic.R')
RRM_op_wtp_age_opt_out <- lapply(variable_levels, function(inc){
print(sprintf('Variable: %s Level: %s', variable, inc))
data.file = 'Part 1 Opt-out.csv'
RRM_boot_opt_out_generic(data.file, seed_nbr=1, variable,
variable_level=inc, p=1,
fix_age = TRUE, split_by_variable = TRUE,
verbose=TRUE)
})## [1] "Variable: age_group Level: 1"
##
## 1 2 3
## 1044 1638 1404
## initial value 1396.709180
## iter 10 value 269.068602
## iter 20 value 244.075410
## final value 244.058570
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00348 0.00039 -8.889 0.00001
## eff_3x:B 2.14818 0.41044 5.234 0.00054
## eff_2x:B 2.58478 0.42878 6.028 0.00020
## eff_1x:B 2.18745 0.41618 5.256 0.00052
## eff_1bix:B 1.79972 0.41131 4.376 0.00178
## dosead1:B 0.06303 0.22321 0.282 0.78406
## sides1:B 0.32471 0.25841 1.257 0.24055
## type1:B 0.15647 0.23603 0.663 0.52398
## beta.optout:B -5.04396 0.18245 -27.646 0.00000
##
## Log-Likelihood: -244.0586
## Iterations: 111.0000
## [1] "Variable: age_group Level: 2"
##
## 1 2 3
## 1044 1638 1404
## initial value 2289.914317
## iter 10 value 639.890834
## final value 356.776574
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00336 0.0003 -11.25 0.00000
## eff_3x:B 1.30452 0.2792 4.67 0.00117
## eff_2x:B 1.46503 0.2813 5.21 0.00056
## eff_1x:B 2.02496 0.2992 6.77 0.00008
## eff_1bix:B 2.24768 0.2943 7.64 0.00003
## dosead1:B 0.42410 0.1837 2.31 0.04629
## sides1:B 0.43193 0.1810 2.39 0.04079
## type1:B -0.20918 0.1823 -1.15 0.28084
## beta.optout:B -5.56791 0.1784 -31.20 0.00000
##
## Log-Likelihood: -356.7766
## Iterations: 86.0000
## [1] "Variable: age_group Level: 3"
##
## 1 2 3
## 1044 1638 1404
## initial value 1966.608226
## iter 10 value 393.357769
## final value 325.947490
## converged
## Estimate Std. Error t-value Pr(>|t|)
## cost:B -0.00256 0.00027 -9.563 0.00001
## eff_3x:B 1.66045 0.30740 5.402 0.00043
## eff_2x:B 1.68981 0.29322 5.763 0.00027
## eff_1x:B 2.14615 0.31629 6.785 0.00008
## eff_1bix:B 2.18528 0.31808 6.870 0.00007
## dosead1:B 0.06238 0.18258 0.342 0.74046
## sides1:B 0.37734 0.18104 2.084 0.06680
## type1:B -0.12870 0.17948 -0.717 0.49152
## beta.optout:B -5.57859 0.19372 -28.797 0.00000
##
## Log-Likelihood: -325.9475
## Iterations: 87.0000names(RRM_op_wtp_age_opt_out) <- variable_levels
save(RRM_op_wtp_age_opt_out, file='RRM_op_wtp_age_opt_out.RData')