Conditional average treatment effects

cate() evaluates treated-minus-control predictive means, or restricted means, at user-supplied covariate rows.

Usage

# S3 method for class 'causalmixgpd_causal_fit'
cate(
  fit,
  newdata = NULL,
  type = c("mean", "rmean"),
  cutoff = NULL,
  interval = "credible",
  level = 0.95,
  nsim_mean = 200L,
  show_progress = TRUE
)

Arguments

fit

A "causalmixgpd_causal_fit" object from run_mcmc_causal().

newdata

Optional data.frame or matrix of covariates for prediction. If NULL, uses the training covariates stored in fit.

type

Character; type of mean treatment effect:

"mean" (default): ordinary mean CATE
"rmean": restricted-mean CATE (requires cutoff)

cutoff

Finite numeric cutoff for restricted mean; required for type = "rmean", ignored otherwise.

interval

Character or NULL; type of credible interval:

NULL: no interval
"credible" (default): equal-tailed quantile intervals
"hpd": highest posterior density intervals

level

Numeric credible level for intervals (default 0.95 for 95 percent CI).

nsim_mean

Number of posterior predictive draws used by simulation-based mean targets. Ignored for analytical ordinary means.

show_progress

Logical; if TRUE, print step messages and render progress where supported.

Value

An object of class "causalmixgpd_ate" containing the CATE summary, optional intervals, and the treated/control prediction objects used to construct the effect. The returned object includes a top-level $fit_df data frame for direct extraction.

Details

For each prediction row $x$, the conditional average treatment effect is $$\mathrm{CATE}(x) = E\{Y(1) \mid x\} - E\{Y(0) \mid x\}.$$

With type = "rmean", the estimand becomes the conditional restricted mean contrast $$E\{\min(Y(1), c) \mid x\} - E\{\min(Y(0), c) \mid x\},$$ which remains finite even when the ordinary mean is unstable under a heavy GPD tail. For outcome kernels with a finite analytical mean, the ordinary mean path is analytical within each posterior draw; rmean remains a separate simulation-based estimand.

This estimand is available only for conditional causal models with covariates. For marginal mean contrasts, use ate or att.

Examples

# \donttest{
N <- 25
X <- data.frame(x1 = stats::rnorm(N))
A <- stats::rbinom(N, 1, 0.5)
y <- abs(stats::rnorm(N)) + 0.1
mcmc_small <- list(niter = 100, nburnin = 50, thin = 1, nchains = 1, seed = 1)
cb <- build_causal_bundle(y = y, X = X, A = A, backend = "sb", kernel = "normal",
                         components = 3, mcmc_outcome = mcmc_small, mcmc_ps = mcmc_small)
fit <- run_mcmc_causal(cb, show_progress = FALSE)
#> Defining model
#> Building model
#> Setting data and initial values
#> Checking model sizes and dimensions
#>   [Note] This model is not fully initialized. This is not an error.
#>          To see which variables are not initialized, use model$initializeInfo().
#>          For more information on model initialization, see help(modelInitialization).
#> ===== Monitors =====
#> thin = 1: beta
#> ===== Samplers =====
#> RW sampler (2)
#>   - beta[]  (2 elements)
#> Compiling
#>   [Note] This may take a minute.
#>   [Note] Use 'showCompilerOutput = TRUE' to see C++ compilation details.
#> Compiling
#>   [Note] This may take a minute.
#>   [Note] Use 'showCompilerOutput = TRUE' to see C++ compilation details.
#> running chain 1...
#> Defining model
#> Building model
#> Setting data and initial values
#> Checking model sizes and dimensions
#>   [Note] This model is not fully initialized. This is not an error.
#>          To see which variables are not initialized, use model$initializeInfo().
#>          For more information on model initialization, see help(modelInitialization).
#> ===== Monitors =====
#> thin = 1: alpha, beta_mean, beta_ps_mean, sd, w
#> ===== Samplers =====
#> conjugate sampler (3)
#>   - sd[]  (3 elements)
#> categorical sampler (15)
#>   - z[]  (15 elements)
#> RW sampler (9)
#>   - alpha
#>   - beta_mean[]  (3 elements)
#>   - beta_ps_mean[]  (3 elements)
#>   - v[]  (2 elements)
#>   [Note] Assigning an RW_block sampler to nodes with very different scales can result in low MCMC efficiency.  If all nodes assigned to RW_block are not on a similar scale, we recommend providing an informed value for the "propCov" control list argument, or using the "barker" sampler instead.
#> Compiling
#>   [Note] This may take a minute.
#>   [Note] Use 'showCompilerOutput = TRUE' to see C++ compilation details.
#> Compiling
#>   [Note] This may take a minute.
#>   [Note] Use 'showCompilerOutput = TRUE' to see C++ compilation details.
#>   [Warning] To calculate WAIC, set 'WAIC = TRUE', in addition to having enabled WAIC in building the MCMC.
#> running chain 1...
#> Defining model
#> Building model
#> Setting data and initial values
#> Checking model sizes and dimensions
#>   [Note] This model is not fully initialized. This is not an error.
#>          To see which variables are not initialized, use model$initializeInfo().
#>          For more information on model initialization, see help(modelInitialization).
#> ===== Monitors =====
#> thin = 1: alpha, beta_mean, beta_ps_mean, sd, w
#> ===== Samplers =====
#> conjugate sampler (3)
#>   - sd[]  (3 elements)
#> categorical sampler (10)
#>   - z[]  (10 elements)
#> RW sampler (9)
#>   - alpha
#>   - beta_mean[]  (3 elements)
#>   - beta_ps_mean[]  (3 elements)
#>   - v[]  (2 elements)
#>   [Note] Assigning an RW_block sampler to nodes with very different scales can result in low MCMC efficiency.  If all nodes assigned to RW_block are not on a similar scale, we recommend providing an informed value for the "propCov" control list argument, or using the "barker" sampler instead.
#> Compiling
#>   [Note] This may take a minute.
#>   [Note] Use 'showCompilerOutput = TRUE' to see C++ compilation details.
#> Compiling
#>   [Note] This may take a minute.
#>   [Note] Use 'showCompilerOutput = TRUE' to see C++ compilation details.
#>   [Warning] To calculate WAIC, set 'WAIC = TRUE', in addition to having enabled WAIC in building the MCMC.
#> running chain 1...
cate(fit, newdata = X[1:5, , drop = FALSE])
#> [cate] Preparing CATE inputs
#> [cate] Preparing propensity-score adjustment
#> [cate] Predicting treated-arm effects
#> [cate] Predicting control-arm effects
#> [cate] Aggregating CATE estimates
#> CATE (Conditional Average Treatment Effect)
#>   Prediction points: 5
#>   Conditional (covariates): YES
#>   Propensity score used: YES
#>   PS scale: logit
#>   Credible interval: credible (95%)
#> 
#> CATE estimates (treated - control):
#>    profile   mean  lower upper
#>  Profile 1   -0.1 -0.348 0.053
#>  Profile 2 -0.089 -0.307 0.047
#>  Profile 3  0.078 -0.041  0.27
#>  Profile 4  0.295 -0.155  1.02
#>  Profile 5  0.062 -0.033 0.215
cate(fit, interval = "credible", level = 0.90)  # 90% CI
#> [cate] Preparing CATE inputs
#> [cate] Preparing propensity-score adjustment
#> [cate] Predicting treated-arm effects
#> [cate] Predicting control-arm effects
#> [cate] Aggregating CATE estimates
#> CATE (Conditional Average Treatment Effect)
#>   Prediction points: 25
#>   Conditional (covariates): YES
#>   Propensity score used: YES
#>   PS scale: logit
#>   Credible interval: credible (90%)
#> 
#> CATE estimates (treated - control):
#>  id   mean  lower upper
#>   1   -0.1 -0.284 0.053
#>   2 -0.089 -0.251 0.047
#>   3  0.078 -0.041 0.221
#>   4  0.295 -0.155 0.834
#>   5  0.062 -0.033 0.176
#>   6 -0.142 -0.403 0.075
#> ... (19 more rows)
cate(fit, interval = "hpd")  # HPD intervals
#> [cate] Preparing CATE inputs
#> [cate] Preparing propensity-score adjustment
#> [cate] Predicting treated-arm effects
#> [cate] Predicting control-arm effects
#> [cate] Aggregating CATE estimates
#> CATE (Conditional Average Treatment Effect)
#>   Prediction points: 25
#>   Conditional (covariates): YES
#>   Propensity score used: YES
#>   PS scale: logit
#>   Credible interval: hpd
#> 
#> CATE estimates (treated - control):
#>  id   mean  lower upper
#>   1   -0.1 -0.366 0.053
#>   2 -0.089 -0.323 0.047
#>   3  0.078 -0.041 0.284
#>   4  0.295 -0.155 1.074
#>   5  0.062 -0.033 0.227
#>   6 -0.142 -0.519 0.075
#> ... (19 more rows)
cate(fit, interval = NULL)   # No intervals
#> [cate] Preparing CATE inputs
#> [cate] Preparing propensity-score adjustment
#> [cate] Predicting treated-arm effects
#> [cate] Predicting control-arm effects
#> [cate] Aggregating CATE estimates
#> CATE (Conditional Average Treatment Effect)
#>   Prediction points: 25
#>   Conditional (covariates): YES
#>   Propensity score used: YES
#>   PS scale: logit
#>   Credible interval: none
#> 
#> CATE estimates (treated - control):
#>  id   mean lower upper
#>   1   -0.1    NA    NA
#>   2 -0.089    NA    NA
#>   3  0.078    NA    NA
#>   4  0.295    NA    NA
#>   5  0.062    NA    NA
#>   6 -0.142    NA    NA
#> ... (19 more rows)
# }

Usage

Arguments

Value

Details

See also

Examples