README.md
In JiayiJi/causal.multiple.treatments: Estimation of Causal Effects of Multiple Treatments with a Binary Outcome
causal.multiple.treatments
Estimation of Causal Effects of Multiple Treatments with a Binary Outcome
The goal of causal.multiple.treatments is to
Installation
You can install the latest version directly from GitHub with devtools:
install.packages("devtools")
devtools::install_github("JiayiJi/causal.multiple.treatments")
Example of use
Data
We generated the data under scenario 1 in the paper.
library(causal.multiple.treatments)
set.seed(3242019)
idata = data_gen(n = 1200, ratio =1)
trt_ind <- as.numeric(idata$trtdat$trt_ind)
all_vars <- idata$trtdat[, -1] #exclude treatment indicator
y <- idata$Yobs
Regression Adjustment
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method ="Regression Adjustment", estimand = "ATE")
## $ATE12
## EST SE LOWER UPPER
## RD -0.026 0.023 -0.073 0.019
## RR 0.918 0.071 0.782 1.060
## OR 0.887 0.099 0.706 1.090
##
## $ATE13
## EST SE LOWER UPPER
## RD -0.054 0.037 -0.124 0.019
## RR 0.850 0.099 0.675 1.060
## OR 0.792 0.136 0.563 1.091
##
## $ATE23
## EST SE LOWER UPPER
## RD -0.027 0.025 -0.074 0.023
## RR 0.925 0.069 0.796 1.073
## OR 0.891 0.100 0.713 1.109
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method ="Regression Adjustment", estimand = "ATT")
## $ATT12
## EST SE LOWER UPPER
## RD -0.027 0.035 -0.094 0.043
## RR 0.920 0.109 0.721 1.142
## OR 0.892 0.151 0.631 1.211
##
## $ATT13
## EST SE LOWER UPPER
## RD -0.054 0.047 -0.151 0.033
## RR 0.852 0.126 0.619 1.122
## OR 0.798 0.174 0.502 1.176
VM Matching
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "VM Matching", estimand = "ATT")
## $ATT12
## EST
## RD -0.08496732
## RR 0.77966102
## OR 0.68493585
##
## $ATT13
## EST
## RD 0.0620915
## RR 1.2602740
## OR 1.3721675
BART with and without discarding
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "BART", estimand = "ATE", discard = TRUE)
## *****Into main of pbart
## *****Data:
## data:n,p,np: 1200, 11, 0
## y1,yn: 0, 1
## x1,x[n*p]: 2.000000, 0.000000
## *****Number of Trees: 100
## *****Number of Cut Points: 2 ... 2
## *****burn and ndpost: 1000, 1000
## *****Prior:mybeta,alpha,tau: 2.000000,0.950000,0.150000
## *****binaryOffset: -0.470030
## *****Dirichlet:sparse,theta,omega,a,b,rho,augment: 0,0,1,0.5,1,11,0
## *****nkeeptrain,nkeeptest,nkeeptreedraws: 1000,1000,1000
## *****printevery: 100
## *****skiptr,skipte,skiptreedraws: 1,1,1
##
## MCMC
## done 0 (out of 2000)
## done 100 (out of 2000)
## done 200 (out of 2000)
## done 300 (out of 2000)
## done 400 (out of 2000)
## done 500 (out of 2000)
## done 600 (out of 2000)
## done 700 (out of 2000)
## done 800 (out of 2000)
## done 900 (out of 2000)
## done 1000 (out of 2000)
## done 1100 (out of 2000)
## done 1200 (out of 2000)
## done 1300 (out of 2000)
## done 1400 (out of 2000)
## done 1500 (out of 2000)
## done 1600 (out of 2000)
## done 1700 (out of 2000)
## done 1800 (out of 2000)
## done 1900 (out of 2000)
## time: 8s
## check counts
## trcnt,tecnt: 1000,0
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## $ATE12
## EST SE LOWER UPPER
## RD 0.002 0.034 -0.064 0.072
## RR 1.008 0.084 0.855 1.191
## OR 1.018 0.145 0.766 1.342
##
## $ATE13
## EST SE LOWER UPPER
## RD -0.061 0.037 -0.137 0.008
## RR 0.875 0.071 0.736 1.020
## OR 0.790 0.117 0.575 1.035
##
## $ATE23
## EST SE LOWER UPPER
## RD -0.063 0.040 -0.144 0.008
## RR 0.872 0.077 0.724 1.018
## OR 0.785 0.125 0.554 1.031
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "BART", estimand = "ATE", discard = FALSE)
## *****Into main of pbart
## *****Data:
## data:n,p,np: 1200, 11, 0
## y1,yn: 0, 1
## x1,x[n*p]: 2.000000, 0.000000
## *****Number of Trees: 100
## *****Number of Cut Points: 2 ... 2
## *****burn and ndpost: 1000, 1000
## *****Prior:mybeta,alpha,tau: 2.000000,0.950000,0.150000
## *****binaryOffset: -0.470030
## *****Dirichlet:sparse,theta,omega,a,b,rho,augment: 0,0,1,0.5,1,11,0
## *****nkeeptrain,nkeeptest,nkeeptreedraws: 1000,1000,1000
## *****printevery: 100
## *****skiptr,skipte,skiptreedraws: 1,1,1
##
## MCMC
## done 0 (out of 2000)
## done 100 (out of 2000)
## done 200 (out of 2000)
## done 300 (out of 2000)
## done 400 (out of 2000)
## done 500 (out of 2000)
## done 600 (out of 2000)
## done 700 (out of 2000)
## done 800 (out of 2000)
## done 900 (out of 2000)
## done 1000 (out of 2000)
## done 1100 (out of 2000)
## done 1200 (out of 2000)
## done 1300 (out of 2000)
## done 1400 (out of 2000)
## done 1500 (out of 2000)
## done 1600 (out of 2000)
## done 1700 (out of 2000)
## done 1800 (out of 2000)
## done 1900 (out of 2000)
## time: 8s
## check counts
## trcnt,tecnt: 1000,0
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## $ATE12
## EST SE LOWER UPPER
## RD -0.002 0.034 -0.070 0.066
## RR 1.000 0.083 0.844 1.177
## OR 1.004 0.143 0.749 1.314
##
## $ATE13
## EST SE LOWER UPPER
## RD -0.055 0.036 -0.124 0.009
## RR 0.885 0.071 0.754 1.021
## OR 0.807 0.116 0.603 1.038
##
## $ATE23
## EST SE LOWER UPPER
## RD -0.054 0.040 -0.129 0.022
## RR 0.889 0.081 0.742 1.051
## OR 0.814 0.134 0.589 1.095
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "BART", estimand = "ATT", discard = TRUE)
## *****Into main of pbart
## *****Data:
## data:n,p,np: 1200, 11, 0
## y1,yn: 0, 1
## x1,x[n*p]: 2.000000, 0.000000
## *****Number of Trees: 100
## *****Number of Cut Points: 2 ... 2
## *****burn and ndpost: 1000, 1000
## *****Prior:mybeta,alpha,tau: 2.000000,0.950000,0.150000
## *****binaryOffset: -0.470030
## *****Dirichlet:sparse,theta,omega,a,b,rho,augment: 0,0,1,0.5,1,11,0
## *****nkeeptrain,nkeeptest,nkeeptreedraws: 1000,1000,1000
## *****printevery: 100
## *****skiptr,skipte,skiptreedraws: 1,1,1
##
## MCMC
## done 0 (out of 2000)
## done 100 (out of 2000)
## done 200 (out of 2000)
## done 300 (out of 2000)
## done 400 (out of 2000)
## done 500 (out of 2000)
## done 600 (out of 2000)
## done 700 (out of 2000)
## done 800 (out of 2000)
## done 900 (out of 2000)
## done 1000 (out of 2000)
## done 1100 (out of 2000)
## done 1200 (out of 2000)
## done 1300 (out of 2000)
## done 1400 (out of 2000)
## done 1500 (out of 2000)
## done 1600 (out of 2000)
## done 1700 (out of 2000)
## done 1800 (out of 2000)
## done 1900 (out of 2000)
## time: 9s
## check counts
## trcnt,tecnt: 1000,0
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## $ATT12
## EST SE LOWER UPPER
## RD 0.003 0.032 -0.061 0.061
## RR 1.014 0.102 0.824 1.212
## OR 1.024 0.151 0.753 1.323
##
## $ATT13
## EST SE LOWER UPPER
## RD -0.133 0.033 -0.199 -0.071
## RR 0.706 0.063 0.589 0.833
## OR 0.571 0.081 0.428 0.741
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "BART", estimand = "ATT", discard = FALSE)
## *****Into main of pbart
## *****Data:
## data:n,p,np: 1200, 11, 0
## y1,yn: 0, 1
## x1,x[n*p]: 2.000000, 0.000000
## *****Number of Trees: 100
## *****Number of Cut Points: 2 ... 2
## *****burn and ndpost: 1000, 1000
## *****Prior:mybeta,alpha,tau: 2.000000,0.950000,0.150000
## *****binaryOffset: -0.470030
## *****Dirichlet:sparse,theta,omega,a,b,rho,augment: 0,0,1,0.5,1,11,0
## *****nkeeptrain,nkeeptest,nkeeptreedraws: 1000,1000,1000
## *****printevery: 100
## *****skiptr,skipte,skiptreedraws: 1,1,1
##
## MCMC
## done 0 (out of 2000)
## done 100 (out of 2000)
## done 200 (out of 2000)
## done 300 (out of 2000)
## done 400 (out of 2000)
## done 500 (out of 2000)
## done 600 (out of 2000)
## done 700 (out of 2000)
## done 800 (out of 2000)
## done 900 (out of 2000)
## done 1000 (out of 2000)
## done 1100 (out of 2000)
## done 1200 (out of 2000)
## done 1300 (out of 2000)
## done 1400 (out of 2000)
## done 1500 (out of 2000)
## done 1600 (out of 2000)
## done 1700 (out of 2000)
## done 1800 (out of 2000)
## done 1900 (out of 2000)
## time: 8s
## check counts
## trcnt,tecnt: 1000,0
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## $ATT12
## EST SE LOWER UPPER
## RD -0.027 0.033 -0.092 0.038
## RR 0.926 0.093 0.757 1.117
## OR 0.895 0.133 0.658 1.185
##
## $ATT13
## EST SE LOWER UPPER
## RD -0.115 0.033 -0.184 -0.054
## RR 0.734 0.067 0.602 0.865
## OR 0.615 0.088 0.449 0.793
IPTW-related methods
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-Logistics", estimand = "ATE")
## $ATE12
## EST
## RD 0.1157188
## RR 1.5143438
## OR 1.7801388
##
## $ATE13
## EST
## RD -0.0601762
## RR 0.8498891
## OR 0.7723171
##
## $ATE23
## EST
## RD -0.1758950
## RR 0.5612260
## OR 0.4338522
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-Logistics-Trim", estimand = "ATE")
## $ATE12
## EST
## RD 0.09896265
## RR 1.44063097
## OR 1.65139284
##
## $ATE13
## EST
## RD -0.07752048
## RR 0.80671882
## OR 0.71426888
##
## $ATE23
## EST
## RD -0.1764831
## RR 0.5599760
## OR 0.4325251
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-GBM", estimand = "ATE")
## $ATE12
## EST
## RD 0.06451888
## RR 1.26600925
## OR 1.38388642
##
## $ATE13
## EST
## RD -0.1142297
## RR 0.7288588
## OR 0.6087075
##
## $ATE23
## EST
## RD -0.1787486
## RR 0.5757137
## OR 0.4398537
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-GBM-Trim", estimand = "ATE")
## $ATE12
## EST
## RD 0.06695134
## RR 1.27088358
## OR 1.39493759
##
## $ATE13
## EST
## RD -0.1087382
## RR 0.7428436
## OR 0.6250761
##
## $ATE23
## EST
## RD -0.1756896
## RR 0.5845096
## OR 0.4481033
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-Logistics", estimand = "ATT")
## $ATT12
## EST
## RD 0.03958638
## RR 1.14995689
## OR 1.21532271
##
## $ATT13
## EST
## RD -0.1299691
## RR 0.7002146
## OR 0.5695390
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-Logistics-Trim", estimand = "ATT")
## $ATT12
## EST
## RD 0.04048331
## RR 1.15387738
## OR 1.22095214
##
## $ATT13
## EST
## RD -0.1302954
## RR 0.6996881
## OR 0.5687829
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-GBM", estimand = "ATT")
## $ATT12
## EST
## RD 0.002632855
## RR 1.008748811
## OR 1.012562395
##
## $ATT13
## EST
## RD -0.1247128
## RR 0.7088083
## OR 0.5818786
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-GBM-Trim", estimand = "ATT")
## $ATT12
## EST
## RD 0.01950579
## RR 1.06866651
## OR 1.09859806
##
## $ATT13
## EST
## RD -0.1215373
## RR 0.7141030
## OR 0.5894812
Contributions
If you find small bugs, larger issues, or have suggestions, please file them using the issue tracker or email the maintainer at Jiayi.Ji@mountsinai.org. Contributions (via pull requests or otherwise) are welcome.
JiayiJi/causal.multiple.treatments documentation built on Nov. 14, 2019, 7:46 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
causal.multiple.treatments
Estimation of Causal Effects of Multiple Treatments with a Binary Outcome
The goal of causal.multiple.treatments is to
Installation
You can install the latest version directly from GitHub with devtools:
install.packages("devtools")
devtools::install_github("JiayiJi/causal.multiple.treatments")
Example of use
Data
We generated the data under scenario 1 in the paper.
library(causal.multiple.treatments)
set.seed(3242019)
idata = data_gen(n = 1200, ratio =1)
trt_ind <- as.numeric(idata$trtdat$trt_ind)
all_vars <- idata$trtdat[, -1] #exclude treatment indicator
y <- idata$Yobs
Regression Adjustment
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method ="Regression Adjustment", estimand = "ATE")
## $ATE12
## EST SE LOWER UPPER
## RD -0.026 0.023 -0.073 0.019
## RR 0.918 0.071 0.782 1.060
## OR 0.887 0.099 0.706 1.090
##
## $ATE13
## EST SE LOWER UPPER
## RD -0.054 0.037 -0.124 0.019
## RR 0.850 0.099 0.675 1.060
## OR 0.792 0.136 0.563 1.091
##
## $ATE23
## EST SE LOWER UPPER
## RD -0.027 0.025 -0.074 0.023
## RR 0.925 0.069 0.796 1.073
## OR 0.891 0.100 0.713 1.109
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method ="Regression Adjustment", estimand = "ATT")
## $ATT12
## EST SE LOWER UPPER
## RD -0.027 0.035 -0.094 0.043
## RR 0.920 0.109 0.721 1.142
## OR 0.892 0.151 0.631 1.211
##
## $ATT13
## EST SE LOWER UPPER
## RD -0.054 0.047 -0.151 0.033
## RR 0.852 0.126 0.619 1.122
## OR 0.798 0.174 0.502 1.176
VM Matching
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "VM Matching", estimand = "ATT")
## $ATT12
## EST
## RD -0.08496732
## RR 0.77966102
## OR 0.68493585
##
## $ATT13
## EST
## RD 0.0620915
## RR 1.2602740
## OR 1.3721675
BART with and without discarding
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "BART", estimand = "ATE", discard = TRUE)
## *****Into main of pbart
## *****Data:
## data:n,p,np: 1200, 11, 0
## y1,yn: 0, 1
## x1,x[n*p]: 2.000000, 0.000000
## *****Number of Trees: 100
## *****Number of Cut Points: 2 ... 2
## *****burn and ndpost: 1000, 1000
## *****Prior:mybeta,alpha,tau: 2.000000,0.950000,0.150000
## *****binaryOffset: -0.470030
## *****Dirichlet:sparse,theta,omega,a,b,rho,augment: 0,0,1,0.5,1,11,0
## *****nkeeptrain,nkeeptest,nkeeptreedraws: 1000,1000,1000
## *****printevery: 100
## *****skiptr,skipte,skiptreedraws: 1,1,1
##
## MCMC
## done 0 (out of 2000)
## done 100 (out of 2000)
## done 200 (out of 2000)
## done 300 (out of 2000)
## done 400 (out of 2000)
## done 500 (out of 2000)
## done 600 (out of 2000)
## done 700 (out of 2000)
## done 800 (out of 2000)
## done 900 (out of 2000)
## done 1000 (out of 2000)
## done 1100 (out of 2000)
## done 1200 (out of 2000)
## done 1300 (out of 2000)
## done 1400 (out of 2000)
## done 1500 (out of 2000)
## done 1600 (out of 2000)
## done 1700 (out of 2000)
## done 1800 (out of 2000)
## done 1900 (out of 2000)
## time: 8s
## check counts
## trcnt,tecnt: 1000,0
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## $ATE12
## EST SE LOWER UPPER
## RD 0.002 0.034 -0.064 0.072
## RR 1.008 0.084 0.855 1.191
## OR 1.018 0.145 0.766 1.342
##
## $ATE13
## EST SE LOWER UPPER
## RD -0.061 0.037 -0.137 0.008
## RR 0.875 0.071 0.736 1.020
## OR 0.790 0.117 0.575 1.035
##
## $ATE23
## EST SE LOWER UPPER
## RD -0.063 0.040 -0.144 0.008
## RR 0.872 0.077 0.724 1.018
## OR 0.785 0.125 0.554 1.031
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "BART", estimand = "ATE", discard = FALSE)
## *****Into main of pbart
## *****Data:
## data:n,p,np: 1200, 11, 0
## y1,yn: 0, 1
## x1,x[n*p]: 2.000000, 0.000000
## *****Number of Trees: 100
## *****Number of Cut Points: 2 ... 2
## *****burn and ndpost: 1000, 1000
## *****Prior:mybeta,alpha,tau: 2.000000,0.950000,0.150000
## *****binaryOffset: -0.470030
## *****Dirichlet:sparse,theta,omega,a,b,rho,augment: 0,0,1,0.5,1,11,0
## *****nkeeptrain,nkeeptest,nkeeptreedraws: 1000,1000,1000
## *****printevery: 100
## *****skiptr,skipte,skiptreedraws: 1,1,1
##
## MCMC
## done 0 (out of 2000)
## done 100 (out of 2000)
## done 200 (out of 2000)
## done 300 (out of 2000)
## done 400 (out of 2000)
## done 500 (out of 2000)
## done 600 (out of 2000)
## done 700 (out of 2000)
## done 800 (out of 2000)
## done 900 (out of 2000)
## done 1000 (out of 2000)
## done 1100 (out of 2000)
## done 1200 (out of 2000)
## done 1300 (out of 2000)
## done 1400 (out of 2000)
## done 1500 (out of 2000)
## done 1600 (out of 2000)
## done 1700 (out of 2000)
## done 1800 (out of 2000)
## done 1900 (out of 2000)
## time: 8s
## check counts
## trcnt,tecnt: 1000,0
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 398
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 410
## ***using serial code
## $ATE12
## EST SE LOWER UPPER
## RD -0.002 0.034 -0.070 0.066
## RR 1.000 0.083 0.844 1.177
## OR 1.004 0.143 0.749 1.314
##
## $ATE13
## EST SE LOWER UPPER
## RD -0.055 0.036 -0.124 0.009
## RR 0.885 0.071 0.754 1.021
## OR 0.807 0.116 0.603 1.038
##
## $ATE23
## EST SE LOWER UPPER
## RD -0.054 0.040 -0.129 0.022
## RR 0.889 0.081 0.742 1.051
## OR 0.814 0.134 0.589 1.095
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "BART", estimand = "ATT", discard = TRUE)
## *****Into main of pbart
## *****Data:
## data:n,p,np: 1200, 11, 0
## y1,yn: 0, 1
## x1,x[n*p]: 2.000000, 0.000000
## *****Number of Trees: 100
## *****Number of Cut Points: 2 ... 2
## *****burn and ndpost: 1000, 1000
## *****Prior:mybeta,alpha,tau: 2.000000,0.950000,0.150000
## *****binaryOffset: -0.470030
## *****Dirichlet:sparse,theta,omega,a,b,rho,augment: 0,0,1,0.5,1,11,0
## *****nkeeptrain,nkeeptest,nkeeptreedraws: 1000,1000,1000
## *****printevery: 100
## *****skiptr,skipte,skiptreedraws: 1,1,1
##
## MCMC
## done 0 (out of 2000)
## done 100 (out of 2000)
## done 200 (out of 2000)
## done 300 (out of 2000)
## done 400 (out of 2000)
## done 500 (out of 2000)
## done 600 (out of 2000)
## done 700 (out of 2000)
## done 800 (out of 2000)
## done 900 (out of 2000)
## done 1000 (out of 2000)
## done 1100 (out of 2000)
## done 1200 (out of 2000)
## done 1300 (out of 2000)
## done 1400 (out of 2000)
## done 1500 (out of 2000)
## done 1600 (out of 2000)
## done 1700 (out of 2000)
## done 1800 (out of 2000)
## done 1900 (out of 2000)
## time: 9s
## check counts
## trcnt,tecnt: 1000,0
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## $ATT12
## EST SE LOWER UPPER
## RD 0.003 0.032 -0.061 0.061
## RR 1.014 0.102 0.824 1.212
## OR 1.024 0.151 0.753 1.323
##
## $ATT13
## EST SE LOWER UPPER
## RD -0.133 0.033 -0.199 -0.071
## RR 0.706 0.063 0.589 0.833
## OR 0.571 0.081 0.428 0.741
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "BART", estimand = "ATT", discard = FALSE)
## *****Into main of pbart
## *****Data:
## data:n,p,np: 1200, 11, 0
## y1,yn: 0, 1
## x1,x[n*p]: 2.000000, 0.000000
## *****Number of Trees: 100
## *****Number of Cut Points: 2 ... 2
## *****burn and ndpost: 1000, 1000
## *****Prior:mybeta,alpha,tau: 2.000000,0.950000,0.150000
## *****binaryOffset: -0.470030
## *****Dirichlet:sparse,theta,omega,a,b,rho,augment: 0,0,1,0.5,1,11,0
## *****nkeeptrain,nkeeptest,nkeeptreedraws: 1000,1000,1000
## *****printevery: 100
## *****skiptr,skipte,skiptreedraws: 1,1,1
##
## MCMC
## done 0 (out of 2000)
## done 100 (out of 2000)
## done 200 (out of 2000)
## done 300 (out of 2000)
## done 400 (out of 2000)
## done 500 (out of 2000)
## done 600 (out of 2000)
## done 700 (out of 2000)
## done 800 (out of 2000)
## done 900 (out of 2000)
## done 1000 (out of 2000)
## done 1100 (out of 2000)
## done 1200 (out of 2000)
## done 1300 (out of 2000)
## done 1400 (out of 2000)
## done 1500 (out of 2000)
## done 1600 (out of 2000)
## done 1700 (out of 2000)
## done 1800 (out of 2000)
## done 1900 (out of 2000)
## time: 8s
## check counts
## trcnt,tecnt: 1000,0
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## *****In main of C++ for bart prediction
## tc (threadcount): 1
## number of bart draws: 1000
## number of trees in bart sum: 100
## number of x columns: 11
## from x,np,p: 11, 392
## ***using serial code
## $ATT12
## EST SE LOWER UPPER
## RD -0.027 0.033 -0.092 0.038
## RR 0.926 0.093 0.757 1.117
## OR 0.895 0.133 0.658 1.185
##
## $ATT13
## EST SE LOWER UPPER
## RD -0.115 0.033 -0.184 -0.054
## RR 0.734 0.067 0.602 0.865
## OR 0.615 0.088 0.449 0.793
IPTW-related methods
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-Logistics", estimand = "ATE")
## $ATE12
## EST
## RD 0.1157188
## RR 1.5143438
## OR 1.7801388
##
## $ATE13
## EST
## RD -0.0601762
## RR 0.8498891
## OR 0.7723171
##
## $ATE23
## EST
## RD -0.1758950
## RR 0.5612260
## OR 0.4338522
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-Logistics-Trim", estimand = "ATE")
## $ATE12
## EST
## RD 0.09896265
## RR 1.44063097
## OR 1.65139284
##
## $ATE13
## EST
## RD -0.07752048
## RR 0.80671882
## OR 0.71426888
##
## $ATE23
## EST
## RD -0.1764831
## RR 0.5599760
## OR 0.4325251
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-GBM", estimand = "ATE")
## $ATE12
## EST
## RD 0.06451888
## RR 1.26600925
## OR 1.38388642
##
## $ATE13
## EST
## RD -0.1142297
## RR 0.7288588
## OR 0.6087075
##
## $ATE23
## EST
## RD -0.1787486
## RR 0.5757137
## OR 0.4398537
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-GBM-Trim", estimand = "ATE")
## $ATE12
## EST
## RD 0.06695134
## RR 1.27088358
## OR 1.39493759
##
## $ATE13
## EST
## RD -0.1087382
## RR 0.7428436
## OR 0.6250761
##
## $ATE23
## EST
## RD -0.1756896
## RR 0.5845096
## OR 0.4481033
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-Logistics", estimand = "ATT")
## $ATT12
## EST
## RD 0.03958638
## RR 1.14995689
## OR 1.21532271
##
## $ATT13
## EST
## RD -0.1299691
## RR 0.7002146
## OR 0.5695390
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-Logistics-Trim", estimand = "ATT")
## $ATT12
## EST
## RD 0.04048331
## RR 1.15387738
## OR 1.22095214
##
## $ATT13
## EST
## RD -0.1302954
## RR 0.6996881
## OR 0.5687829
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-GBM", estimand = "ATT")
## $ATT12
## EST
## RD 0.002632855
## RR 1.008748811
## OR 1.012562395
##
## $ATT13
## EST
## RD -0.1247128
## RR 0.7088083
## OR 0.5818786
causal_multiple_treatment(y = y, x = idata$trtdat, trt = trt_ind, method = "IPTW-GBM-Trim", estimand = "ATT")
## $ATT12
## EST
## RD 0.01950579
## RR 1.06866651
## OR 1.09859806
##
## $ATT13
## EST
## RD -0.1215373
## RR 0.7141030
## OR 0.5894812
Contributions
If you find small bugs, larger issues, or have suggestions, please file them using the issue tracker or email the maintainer at Jiayi.Ji@mountsinai.org. Contributions (via pull requests or otherwise) are welcome.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.