simulate_cosinor function

Simulate data from a cosinor model

This function simulates data from a cosinor model with a single covariate, where the time scale is month, and optionally allows for single covariate effects on the mean, amplitude, and acrophase.

simulate_cosinor(
  n,
  mesor,
  amp,
  acro,
  period = 24,
  n_components,
  beta.group = FALSE,
  beta.mesor,
  beta.amp,
  beta.acro,
  n_period = 1,
  family = c("gaussian", "poisson", "binomial", "gamma"),
  ...
)

Arguments

• n: The sample size. An integer greater than 0.

• mesor: A numeric. The MESOR (midline estimating statistic of rhythm) for group = 0. The MESOR is independent of the cosinor components, so only one value is allowed even if there are multiple components in the data being simulated.

• amp: A numeric. The amplitude value (for group = 0 if grouped data are being simulated (beta.group = TRUE)). If simulating data with multiple components, specify a vector with values for each component. E.g: amp = c(5, 10).

• acro: A numeric. The acrophase value in radians (for group = 0 if grouped data are being simulated (beta.group = TRUE)). If simulating data with multiple components, specify a vector with values for each component. E.g: acr = c(0, pi)

  for two components.

• period: The period of the rhythm data (for group = 0 if grouped data are being simulated (beta.group = TRUE)). If simulating data with multiple components, specify a vector with values for each component. E.g: period = c(12, 6) for two components.

• n_components: The number of components in the model. This must match the length of the inputs for amp and acro.

• beta.group: A logical. If TRUE a second group of data will be simulated and included in the returned data set. If FALSE, beta.acro, beta.mesor, and beta.amp arguments will be ignored.

• beta.mesor: A numeric. The MESOR value term for group = 1

• beta.amp: A numeric. The amplitude value for group = 1. If simulating data with multiple components, specify a vector with values for each component. E.g: amp = c(2, 8).

• beta.acro: A numeric. The acrophase value in radians (for group = 1. If simulating data with multiple components, specify a vector with values for each component. E.g: acr = c(2, 5)

  for two components.

• n_period: A numeric. The number of cycles of the rhythm to be simulated.

• family: A character. The family (see ?family) of the simulated dataset. Can handle values in c("poisson", "binomial", "gamma", "gaussian").

• ...: Extra arguments, including alpha that controls the shape argument when sampling from a gamma distribution (when family = "gamma"; default is 1), and sd

  (standard deviation) which is used when sampling from a normal distribution (when family = "gaussian"; default is 1). To specify these parameters for the beta (treatment) group, use beta.alpha and beta.sd

Returns

Returns simulated data in a data.frame.

Examples

simulate_cosinor(
  n = 100,
  mesor = 1,
  amp = 1,
  acro = 1,
  period = 24,
  family = "gaussian"
)