Returns structured diagnostics for intervention support, bootstrap behavior,
and MSM adequacy from a fitted qgcomp.glm.multi() object.
Usage
diagnostics(object, type = c("all", "support", "bootstrap", "adequacy"), ...)Details
These diagnostics are deliberately kept outside the main fitted-model
summary so that users can inspect model fit, support, and computational
behavior explicitly rather than having a very dense summary() method.
The three main diagnostic families are:
Support diagnostics, which summarize the fit-time intervention grid. For
q = NULL, these diagnostics are especially important because the intervention grid is built from pooled percentile values within each mixture, and under each intervention every component in a mixture is set to the same pooled mixture-specific value.Bootstrap diagnostics, which summarize how many bootstrap replications were requested, retained, or failed, together with any lightweight failure metadata stored in the fitted object.
MSM adequacy diagnostics, which compare the exact fit-time counterfactual surface to the fitted MSM surface. For transformed-scale fits, the primary adequacy comparison is made on the MSM fitting scale.
Support diagnostics should not be read as a full positivity proof. They are designed to help users see which intervention values define the stored surface and to highlight when original-scale pooled interventions may deserve extra scrutiny.
MSM adequacy is one of the most method-specific diagnostics in the package. It asks whether the fitted MSM is a reasonable low-dimensional summary of the exact fit-time surface implied by the fitted outcome model. It is not a test of whether the outcome model is true, and it is not a general check of causal identification.
Adequacy is evaluated on the stored fit-time grid. A good adequacy result therefore means that the MSM tracks the exact fit-time surface well on that grid. It does not imply that the surface is globally linear between or beyond those intervention points.
Examples
if (FALSE) { # \dontrun{
dat <- sim_mixture_data(
n = 400,
pA = 3,
pB = 3,
rho_within_A = 0.3,
rho_within_B = 0.3,
rho_between = 0.2,
psi1 = 0.5,
psi2 = 0.3,
psi12 = 0.2,
seed = 123
)
fit <- qgcomp.glm.multi(
f = Y ~ X1 + X2 + X3 + W1 + W2 + W3 + C,
data = dat,
mix1 = c("X1", "X2", "X3"),
mix2 = c("W1", "W2", "W3"),
q = 4,
B = 100,
seed = 13
)
# Full diagnostic bundle
diagnostics(fit)
# Focused diagnostics
support(fit)
diagnostics(fit, type = "bootstrap")
adequacy(fit)
} # }