simulate_functional_data: Simulate functional data

In registr: Curve Registration for Exponential Family Functional Data

Simulate functional data

Description

This function simulates functional data. The data it outputs is generated from a mean function and two orthogonal principal component basis functions. The mean and principal components are based on sine and cosine functions. Subject-specific scores for each PC are drawn from normal distributions with standard deviation lambda1 and lambda2.

Usage

simulate_functional_data(
  lambda1 = 2,
  lambda2 = 1,
  I = 50,
  D = 100,
  seed = 1988,
  vary_D = FALSE
)

Arguments

lambda1	Standard deviation for PC1 scores.
lambda2	Standard deviation for PC2 scores.
I	Number of subjects. Defaults is 50.
D	Number of grid points per subject. Default is 100.
seed	Seed for reproducibility. Default is 1988.
vary_D	Indicates if grid length vary by subject. If FALSE all subjects have grid length D.

Value

A list containing:

Y	Simulated dataframe with variables id, value, index, and latent_mean.
psi1	True values for first principal component.
psi2	True values for second principal component.
alpha	True values for population-level mean.

A list containing:

Y	A dataframe of simulated data.
psi1	The first simulated eigenfunction.
psi2	The second simulated eigenfunction.
alpha	The population mean.

Author(s)

Julia Wrobel [e-mail]

registr documentation built on Oct. 3, 2022, 1:05 a.m.