Package index
Fitting
The two models. They take the same data and the same formula, so a model can be refitted under the other by changing the function name alone.
-
pcdreg() - Semiparametric regression for panel count data
Specifying the data
The response constructor, used on the left hand side of the formula. See the data preparation article for the layouts it accepts.
-
pcd() - Describe panel count observations
-
is.pcd() - Test for a panel count response
-
autoplot(<pcd>)plot(<pcd>) - Plot panel count data
Inspecting a fit
Both fitters return an object inheriting from pcdfit, so these work the same way for either.
-
vcov(<pcdfit>) - Covariance matrix of the estimated coefficients
-
summary(<pcdfit>) - Summarise a fitted panel count model
-
baseline() - Estimated baseline function
-
predict(<pcdfit>) - Predictions from a panel count model
-
autoplot(<pcdfit>)plot(<pcdfit>) - Plot the estimated baseline function
Tidy output
Methods for the broom generics, so a fit drops into a tidy workflow: one row per coefficient, a one row summary of the fit, and the data with fitted means and residuals attached.
-
tidy(<pcdfit>)glance(<pcdfit>)augment(<pcdfit>) - Tidy a panel count model fit
-
reexportsautoplottidyglanceaugment - Objects exported from other packages
-
pcdreg_control() - Tuning parameters for the fitting algorithms
Simulation
The design of the simulation study in Sun et al. (2024), useful for examples and for checking the covariance estimators against known truth.
-
sim_pcd() - Simulate panel count data with a time-varying covariate
-
pcdreg-package - pcdreg: Semiparametric Regression for Panel Count Data