| PSweight {PSweight} | R Documentation |
Estimate average causal effects by propensity score weighting
Description
The function PSweight is used to estimate the average potential outcomes corresponding to
each treatment group among the target population. The function currently implements
the following types of weights: the inverse probability of treatment weights (IPW: target population is the combined population),
average treatment effect among the treated weights (treated: target population is the population receiving a specified treatment),
overlap weights (overlap: target population is the overlap population at clinical equipoise), matching weights (matching: target population
is population obtained under 1:1 matching), entropy weights (entropy: target population is the population weighted by the entropy function).
Augmented propensity score weighting estimators are also allowed, with propensity scores and outcome model estimates either estimated
within the function, or supplied by external routines.
Usage
PSweight(
ps.formula = NULL,
ps.estimate = NULL,
trtgrp = NULL,
zname = NULL,
yname,
data,
weight = "overlap",
delta = 0,
augmentation = FALSE,
bootstrap = FALSE,
R = 50,
out.formula = NULL,
out.estimate = NULL,
family = "gaussian",
ps.method = "glm",
ps.control = list(),
out.method = "glm",
out.control = list()
)
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 |
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
and outcome model (if augmented estimator is used). If not found in data, the variables are
taken from |
weight |
a character or vector of characters including the types of weights to be used.
|
delta |
trimming threshold for estimated (generalized) propensity scores. Should be no larger than 1 / number of treatment groups. Default is 0, corresponding to no trimming. |
augmentation |
logical. Indicate whether augmented weighting estimators should be used.
Default is |
bootstrap |
logical. Indaicate whether bootstrap is used to estimate the standard error
of the point estimates. Default is |
R |
an optional integer indicating number of bootstrap replicates. Default is |
out.formula |
an object of class |
out.estimate |
an optional matrix or data frame containing estimated potential outcomes
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, with a
similar mechanism as in |
family |
a description of the error distribution and link function to be used in the outcome model.
Only required if |
ps.method |
a character to specify the method for estimating propensity scores. |
ps.control |
a list to specify additional options when |
out.method |
a character to specify the method for estimating the outcome regression model. |
out.control |
a list to specify additional options 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. Similarly, a typical form for out.formula is
outcome ~ terms where outcome is the outcome variable (identical to the variable name
used to specify yname) and terms is a series of terms which specifies a linear
predictor for outcome. Both ps.formula and out.formula by default specify generalized
linear models when ps.estimate and/or out.estimate is NULL. The argument ps.method and out.method allow user to choose
model other than glm to fit the propensity score and outcome regression models for augmentation. Additional argument in the gbm() function can be supplied through the ps.control and out.control argument. Please refer to the user manual of the gbm package for all the
allowed arguments. "SuperLearner" is also allowed in the ps.method and out.method arguments. Currently, the SuperLearner method only supports binary treatment with the default method set to "SL.glm". The estimation approach is default to "method.NNLS" for both propensity and outcome regression models.
Prediction algorithm and other tuning parameters can also be passed throughps.control and out.control. Please refer to the user manual of the SuperLearner package for all the allowed specifications.
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.
When comparing multiple (J>=3) treatments, ps.estimate needs to be specified as an N by J matrix,
where N indicates the number of observations, and J indicates the total number of treatments.
This matrix specifies the estimated generalized propensity scores to receive each of the J treatments.
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 same mechanism applies to out.estimate, except that the input for out.estimate
must be an N by J matrix, where each row corresponds to the estimated potential outcomes (corresponding to each treatment)
for each observation.
The argument zname and/or yname is required when ps.estimate
and/or out.estimate is not NULL.
Current version of PSweight allows for five types of propensity score weights used to estimate ATE (IPW), ATT (treated) and
ATO (overlap), ATM (matching) and ATEN (entropy). These weights are members of larger class of balancing weights defined in Li, Morgan, and Zaslavsky (2018).
Specific definitions of these weights are provided in Li, Morgan, and Zaslavsky (2018), Li and Greene (2013), Zhou, Matsouaka and Thomas (2020).
When there is a practical violation of the positivity assumption, delta defines the symmetric
propensity score trimming rule following Crump et al. (2009). With multiple treatments, delta defines the
multinomial trimming rule introduced in Yoshida et al. (2019). 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).
If augmentation = TRUE, an augmented weighting estimator will be implemented. For binary treatments, the augmented
weighting estimator is presented in Mao, Li and Greene (2018). For multiple treatments, the augmented weighting estimator is
mentioned in Li and Li (2019), and additional details will appear in our ongoing work (Zhou et al. 2020+). When
weight = "IPW", the augmented estimator is also referred to as a doubly-robust (DR) estimator.
When bootstrap = TRUE, the variance will be calculated by nonparametric bootstrap, with R bootstrap
replications. The default of R is 50. Otherwise, the variance will be calculated using the sandwich variance
formula obtained in the M-estimation framework.
Value
PSweight returns a PSweight object containing a list of the following values:
estimated propensity scores, average potential outcomes corresponding to each treatment,
variance-covariance matrix of the point estimates, the label for each treatment group,
and estimates in each bootstrap replicate if bootstrap = TRUE.
A summary of PSweight can be obtained with summary.PSweight.
trtgrpa character indicating the treatment group.
propensitya data frame of estimated propensity scores.
muhataverage potential outcomes by treatment groups, with reference to specific target populations.
covmuvariance-covariance matrix of
muhat.mubootan optional list of point estimates in each bootstrap replicate
bootstrap = TRUE.groupa table of treatment group labels corresponding to the output point estimates
muhat.
References
Crump, R. K., Hotz, V. J., Imbens, G. W., Mitnik, O. A. (2009). Dealing with limited overlap in estimation of average treatment effects. Biometrika, 96(1), 187-199.
Li, L., Greene, T. (2013). A weighting analogue to pair matching in propensity score analysis. The International Journal of Biostatistics, 9(2), 215-234.
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.
Yoshida, K., Solomon, D.H., Haneuse, S., Kim, S.C., Patorno, E., Tedeschi, S.K., Lyu, H., Franklin, J.M., Stürmer, T., Hernández-Díaz, S. and Glynn, R.J. (2019). Multinomial extension of propensity score trimming methods: A simulation study. American Journal of Epidemiology, 188(3), 609-616.
Li, F., Li, F. (2019). Propensity score weighting for causal inference with multiple treatments. The Annals of Applied Statistics, 13(4), 2389-2415.
Zhou, Y., Matsouaka, R. A., Thomas, L. (2020). Propensity score weighting under limited overlap and model misspecification. Statistical Methods in Medical Research, 29(12), 3721-3756.
Examples
data("psdata")
# the propensity and outcome models
ps.formula<-trt~cov1+cov2+cov3+cov4+cov5+cov6
out.formula<-Y~cov1+cov2+cov3+cov4+cov5+cov6
# without augmentation
ato1<-PSweight(ps.formula = ps.formula,yname = 'Y',data = psdata,weight = 'overlap')
summary(ato1)
# augmented weighting estimator, takes longer time to calculate sandwich variance
# ato2<-PSweight(ps.formula = ps.formula,yname = 'Y',data = psdata,
# augmentation = TRUE,out.formula = out.formula,family = 'gaussian',weight = 'overlap')
# summary(ato2)