brokenstick-class: Class `brokenstick`

Description

The main fitting function brokenstick() returns an object of class brokenstick. This object collects the fitted broken stick model.

names: A list with elements named x, y and g providing the variables names for the time, outcome and subject columns, respectively.
knots: Numeric vector of with the internal knots. Use get_knots() to extract knots.
boundary: Numeric vector of length 2 with the boundary knots. Use get_knots() to extract knots.
degree: The degree of the B-spline. See splines::bs(). Either 0 (constant model) or 1 (broken stick model).
model: Not used.
method: Either lmer or kr, identifying the fitting model.
control: List of control option returned by control_brokenstick().
beta: Numeric vector with fixed effect estimates.
omega: Numeric matrix with variance-covariance estimates of the random effect.
sigma2j: Numeric vector with estimates of the residual variance per group. Only used by method kr.
sigma2: Numeric scalar with the mean residual variance.
draws: Numeric matrix with multiple imputations. The number of rows is equal to the number of missing values in y. The number of columns depends on imp_skip. Only used by kr if imp_skip is set.

The package exports S3 methods for the brokenstick class for the following generic functions: fitted(), model.frame(), model.matrix(), plot(), predict(), print(), residuals().

The package documents methods fitted.brokenstick(), plot.brokenstick(), predict.brokenstick() and residuals.brokenstick(). The package exports two helper functions for brokenstick objects: get_knots() and get_r2().

A brokenstick object is a list with the following named elements:

van Buuren S (2020). Broken Stick Model for Irregular Longitudinal Data. In preparation.