Skip to contents

Estimate measurement noise with constant variance

Source: R/model_measurements.R
noise_estimate_constant_var.Rd

This option estimates the unexplained variation in wastewater measurements using a constant variance model. This is usually a misspecification and is only supported for comparison purposes.

For a constant coefficient of variation model, see noise_estimate(), and for a non-constant coefficient of variation model, see noise_estimate_dPCR().

If multiple measurements (replicates) per sample are provided, EpiSewer can also explicitly model variation before the replication stage.

Usage

noise_estimate_constant_var(
  replicates = FALSE,
  cv_prior_mu = 0,
  cv_prior_sigma = 1,
  pre_replicate_cv_prior_mu = 0,
  pre_replicate_cv_prior_sigma = 1,
  warn = TRUE,
  modeldata = modeldata_init()
)

Arguments

replicates

Should replicates be used to explicitly model variation before the replication stage?

cv_prior_mu

Prior (mean) on the coefficient of variation of concentration measurements. Note that when replicates=TRUE, this is only the CV after the replication stage (see details for more explanation).

cv_prior_sigma

Prior (standard deviation) on the coefficient of variation of concentration measurements.

pre_replicate_cv_prior_mu

Prior (mean) on the coefficient of variation of concentrations before the replication stage.

pre_replicate_cv_prior_sigma

Prior (standard deviation) on the coefficient of variation of concentrations before the replication stage.

modeldata

A modeldata object to which the above model specifications should be added. Default is an empty model given by modeldata_init(). Can also be an already partly specified model returned by other EpiSewer modeling functions.

Value

A modeldata object containing data and specifications of the model to be fitted. Can be passed on to other EpiSewer modeling functions to add further data and model specifications.

The modeldata object also includes information about parameter initialization (.init), meta data (.metainfo), and checks to be performed before model fitting (.checks).

Details

Note that although this model keeps the variance constant, the prior for the measurement noise is still in terms of the (average) coefficient of variation (CV). This makes prior specification easier since the CV is unitless.

The priors of this component have the following functional form:

  • coefficient of variation of concentration measurements: Truncated normal

  • coefficient of variation of concentration before the replication stage: Truncated normal

See also

Other noise models: noise_estimate(), noise_estimate_dPCR()