PSweight_sga: Estimate subgroup causal effects by propensity score...
In siyunyang/PSweight.sga: PSweight.sga: Propensity Score Weighting for Causal Subgroup Analysis
PSweight_sga R Documentation
Estimate subgroup causal effects by propensity score weighting
Description
The function PSweight_sga is used to estimate the subgroup average potential outcomes corresponding to
each treatment group among the target population. The function currently implements
three types of weights: the inverse probability weights (target population is the combined population),
ATT weights (target population is the population receiving one treatment) and overlap weights (target
population is the overlap population at clinical equipoise).
Usage
PSweight_sga(
ps.formula = NULL,
ps.estimate = NULL,
subgroup = NULL,
xname = NULL,
trtgrp = NULL,
zname = NULL,
yname,
data,
R = 50,
weight = "overlap",
method = "glm"
)
Arguments
ps.formula
an object of class formula (or one that can be coerced to that class):
a symbolic description of the propensity score model to be fitted. Additional details of model specification
are given under "Details". This argument is optional if ps.estimate is not NULL.
ps.estimate
an optional matrix or data frame containing estimated (generalized) propensity scores for
each observation. Typically, this is an N by J matrix, where N is the number of observations and J is the
total number of treatment levels. Preferably, the column name of this matrix should match the name of treatment level,
if column name is missing or there is a mismatch, the column names would be assigned according to alphabatic order
of the treatment levels. A vector of propensity score estimates is also allowed in ps.estimate, in which
case a binary treatment is implied and the input is regarded as the propensity to receive the last category of
treatment by alphabatic order, unless otherwise stated by trtgrp.
subgroup
a vector to specify name of subgroup variables by column index or column names
xname
an optional character vector specifying the name of the covariates (confounders) in data. Only continuous and factors are accepted.
trtgrp
an optional character defining the "treated" population for estimating the average treatment
effect among the treated (ATT). Only necessary if weight = "ATT". This option can also be used to specify
the treatment (in a two-treatment setting) when a vector argument is supplied for ps.estimate.
Default value is the last group in the alphebatic order.
zname
an optional character specifying the name of the treatment variable in data.
yname
an optional character specifying the name of the outcome variable in data.
data
an optional data frame containing the variables in the propensity score model.
R
an optional integer indicating number of bootstrap replicates. Default is R = 50.
weight
a character or vector of characters including the types of weights to be used.
"ATE" specifies the inverse probability weights for estimating the average treatment
effect among the combined population. "ATT" specifies the weights for estimating the
average treatment effect among the treated. "ATO" specifies the (generalized) overlap weights
for estimating the average treatment effect among the overlap population, or population at
clinical equipoise. Default is "ATO".
method
a character to specify the method for propensity model. When ps.formula is given, "glm" is the default; When xname is given, "LASSO" is the default.
Details
A typical form for ps.formula is treatment ~ terms where treatment is the treatment
variable (identical to the variable name used to specify zname) and terms is a series of terms
which specifies a linear predictor for treatment. The ps.formula specifies generalized
linear models when ps.estimate is NULL. See glm for more details on generalized linear models.
In Yang et al.(2021), the term is suggested to include all main effects and pairwise interactions between subgroup variables
and confounders. Then LASSO will be performed to select important interactions, and re-fit a logistic regression with the main
effects and LASSO selected interactions (pLASSO). If xname, zname and subgroup are provided, the function will automatically perform pLASSO, and the ps.formula is not required.
When comparing two treatments, ps.estimate can either be a vector or a two-column matrix of estimated
propensity scores. If a vector is supplied, it is assumed to be the propensity scores to receive the treatment, and
the treatment group corresponds to the last group in the alphebatic order, unless otherwise specified by trtgrp.
In general, ps.estimate should have column names that indicate the level of the treatment variable,
which should match the levels given in Z.
If column names are empty or there is a mismatch, the column names will be created following
the alphebatic order of values in Z, and the rightmost coulmn of ps.estimate is assumed
to be the treatment group, when estimating ATT. trtgrp can also be used to specify the treatment
group for estimating ATT.
The argument zname and/or xname is required when ps.estimate is not NULL.
Current version of PSweight_sga allows for three types of propensity score weights used to estimate ATE, ATT and
ATO. These weights are members of larger class of balancing weights defined in Li, Morgan, and Zaslavsky (2018). The overlap weights can also be considered as
a data-driven continuous trimming strategy without specifying trimming rules, see Li, Thomas and Li (2019).
Additional details on balancing weights and generalized overlap weights for multiple treatment groups are provided in
Li and Li (2019).
The variance will be calculated by nonparametric bootstrap, with R bootstrap
replications. The default of R is 50.
Value
PSweight_sga returns a PSweight_sga object containing a list of the following values:
estimated propensity scores, average subgroup potential outcomes corresponding to each treatment,
estimates in each bootstrap replicate, the label for each treatment group, and the number of subgroup
levels defined by each subgrouping variable .
A summary of PSweight_sga can be obtained with summary.PSweight_sga.
propensitya data frame of estimated propensity scores.
muhat average subgroup potential outcomes by treatment groups, with reference to specific target populations.
mubootlist of point estimates in each bootstrap replicate.
group a table of treatment group labels corresponding to the output point estimates muhat.
trtgrpa character indicating the treatment group.
sub_na vector indicating the number of subgroup levels defined by each subgrouping variable .
methoda character indicating the propensity score method used.
References
Yang, S., Lorenzi, E., Papadogeorgou, G., Wojdyla, D. M., Li, F., & Thomas, L. E. (2021).
Propensity score weighting for causal subgroup analysis. Statistics in medicine, 40(19), 4294-4309.
Li, F., Morgan, K. L., Zaslavsky, A. M. (2018).
Balancing covariates via propensity score weighting.
Journal of the American Statistical Association, 113(521), 390-400.
Mao, H., Li, L., Greene, T. (2019). Propensity score weighting analysis and treatment effect discovery.
Statistical Methods in Medical Research, 28(8), 2439-2454.
Li, F., Thomas, L. E., Li, F. (2019).
Addressing extreme propensity scores via the overlap weights. American Journal of Epidemiology, 188(1), 250-257.
siyunyang/PSweight.sga documentation built on Aug. 16, 2022, 5:23 a.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.
| PSweight_sga | R Documentation |
Estimate subgroup causal effects by propensity score weighting
Description
The function PSweight_sga is used to estimate the subgroup average potential outcomes corresponding to
each treatment group among the target population. The function currently implements
three types of weights: the inverse probability weights (target population is the combined population),
ATT weights (target population is the population receiving one treatment) and overlap weights (target
population is the overlap population at clinical equipoise).
Usage
PSweight_sga( ps.formula = NULL, ps.estimate = NULL, subgroup = NULL, xname = NULL, trtgrp = NULL, zname = NULL, yname, data, R = 50, weight = "overlap", method = "glm" )
Arguments
ps.formula |
an object of class |
ps.estimate |
an optional matrix or data frame containing estimated (generalized) propensity scores for
each observation. Typically, this is an N by J matrix, where N is the number of observations and J is the
total number of treatment levels. Preferably, the column name of this matrix should match the name of treatment level,
if column name is missing or there is a mismatch, the column names would be assigned according to alphabatic order
of the treatment levels. A vector of propensity score estimates is also allowed in |
subgroup |
a vector to specify name of subgroup variables by column index or column names |
xname |
an optional character vector specifying the name of the covariates (confounders) in |
trtgrp |
an optional character defining the "treated" population for estimating the average treatment
effect among the treated (ATT). Only necessary if |
zname |
an optional character specifying the name of the treatment variable in |
yname |
an optional character specifying the name of the outcome variable in |
data |
an optional data frame containing the variables in the propensity score model. |
R |
an optional integer indicating number of bootstrap replicates. Default is |
weight |
a character or vector of characters including the types of weights to be used.
|
method |
a character to specify the method for propensity model. When |
Details
A typical form for ps.formula is treatment ~ terms where treatment is the treatment
variable (identical to the variable name used to specify zname) and terms is a series of terms
which specifies a linear predictor for treatment. The ps.formula specifies generalized
linear models when ps.estimate is NULL. See glm for more details on generalized linear models.
In Yang et al.(2021), the term is suggested to include all main effects and pairwise interactions between subgroup variables
and confounders. Then LASSO will be performed to select important interactions, and re-fit a logistic regression with the main
effects and LASSO selected interactions (pLASSO). If xname, zname and subgroup are provided, the function will automatically perform pLASSO, and the ps.formula is not required.
When comparing two treatments, ps.estimate can either be a vector or a two-column matrix of estimated
propensity scores. If a vector is supplied, it is assumed to be the propensity scores to receive the treatment, and
the treatment group corresponds to the last group in the alphebatic order, unless otherwise specified by trtgrp.
In general, ps.estimate should have column names that indicate the level of the treatment variable,
which should match the levels given in Z.
If column names are empty or there is a mismatch, the column names will be created following
the alphebatic order of values in Z, and the rightmost coulmn of ps.estimate is assumed
to be the treatment group, when estimating ATT. trtgrp can also be used to specify the treatment
group for estimating ATT.
The argument zname and/or xname is required when ps.estimate is not NULL.
Current version of PSweight_sga allows for three types of propensity score weights used to estimate ATE, ATT and
ATO. These weights are members of larger class of balancing weights defined in Li, Morgan, and Zaslavsky (2018). The overlap weights can also be considered as
a data-driven continuous trimming strategy without specifying trimming rules, see Li, Thomas and Li (2019).
Additional details on balancing weights and generalized overlap weights for multiple treatment groups are provided in
Li and Li (2019).
The variance will be calculated by nonparametric bootstrap, with R bootstrap
replications. The default of R is 50.
Value
PSweight_sga returns a PSweight_sga object containing a list of the following values:
estimated propensity scores, average subgroup potential outcomes corresponding to each treatment,
estimates in each bootstrap replicate, the label for each treatment group, and the number of subgroup
levels defined by each subgrouping variable .
A summary of PSweight_sga can be obtained with summary.PSweight_sga.
propensitya data frame of estimated propensity scores.
muhataverage subgroup potential outcomes by treatment groups, with reference to specific target populations.
mubootlist of point estimates in each bootstrap replicate.
groupa table of treatment group labels corresponding to the output point estimates
muhat.trtgrpa character indicating the treatment group.
sub_na vector indicating the number of subgroup levels defined by each subgrouping variable .
methoda character indicating the propensity score method used.
References
Yang, S., Lorenzi, E., Papadogeorgou, G., Wojdyla, D. M., Li, F., & Thomas, L. E. (2021). Propensity score weighting for causal subgroup analysis. Statistics in medicine, 40(19), 4294-4309.
Li, F., Morgan, K. L., Zaslavsky, A. M. (2018). Balancing covariates via propensity score weighting. Journal of the American Statistical Association, 113(521), 390-400.
Mao, H., Li, L., Greene, T. (2019). Propensity score weighting analysis and treatment effect discovery. Statistical Methods in Medical Research, 28(8), 2439-2454.
Li, F., Thomas, L. E., Li, F. (2019). Addressing extreme propensity scores via the overlap weights. American Journal of Epidemiology, 188(1), 250-257.
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.