generate_four_param_cov_matrix: Generates covariance matrix for nonlinear longitudinal data...

In sciarraseb/nonlinSims: Generates longitudinal data sets according to a logistic pattern.

Generates covariance matrix for nonlinear longitudinal data that follow logistic pattern.

Description

Four parameters (fixed effects) are used to characterize the logistic pattern and, due to the hierarchical nature of the to-be-generated data, each parameter has a corresponding value of variability (i.e., random-effect). The random effects are used to generate the covariance matrix. Note that correlations between random effects must also be set and that correlations between random effects and error variance at each time point are set to 0 by default (cor_param_error = 0). Internally, the function also assumes zero-value correlations between error variances at each time point The four parameters that characterize the logistic pattern of change take on the following meanings:

• diff: different between first and last values (i.e., difference between two plateaus)

• beta: amount of time to reach midway point (i.e., 50% of the distance between theta and alpha) from time = 0

• gamma: amount of time to reach satiation point (i.e., 73% of distance between theta and alpha) from midpoint

Usage

generate_four_param_cov_matrix(num_time_points, pop_param_list)

Arguments

num_time_points	number of time points
pop_param_list	list of population parameters returned from generate_pop_param_list

Value

Returns a covariance matrix.

sciarraseb/nonlinSims documentation built on Jan. 30, 2023, 8:17 p.m.