Fitted parameter estimates from a wisp model

Returns the fitted wisp model parameter estimates as a plain named numeric vector, following the usual R convention for coef() methods.

Usage

# S3 method for class 'wisp'
coef(object, ...)

Arguments

object: List, output of the wisp function (an object of class "wisp").
...: Ignored; present only for S3 method consistency with coef.

Value

A named numeric vector of fitted parameter estimates (names are the parameter names, matching object$param.names).

Note

object$fitted.parameters is used directly, rather than object$stats$parameters$estimate, as the source: it's already a named numeric vector (no extra reshaping needed to meet the coef() convention), and – unlike object$stats$parameters$estimate – it's always present, even for a model fit with fit_only = TRUE where no statistical analysis was ever run. The two are otherwise identical.

The returned vector mixes several different kinds of quantities, distinguishable by name (matching object$param.names; see model.component.list in the wisp return value for the three base component names). "Rt" entries are the fitted rate on the log(sum + 1) scale of the summed count within a bin, not the original per-instance rate – to recover the latter, exponentiate the value, subtract one, and divide by the mean number of data rows per bin (see the wisp examples, or the "Poisson processes and sigmoids" tutorial, for the full back-transformation). "tslope" and "tpoint" entries are already on their natural, directly interpretable scales and need no such transformation. Whenever the random-effect grouping variable (ran) has more than one level, the vector additionally contains "wfactor_rate", "wfactor_slope", and "wfactor_point" entries (one per random-effect level) – these are warping factors feeding into the warp function described in the "Random effects and warping" tutorial, a different kind of quantity altogether from the like-named base parameters; neither the "Rt" back-transformation above nor "use as-is" applies to them.

Examples

set.seed(1)
n_bins <- 20
bin <- sample(seq_len(n_bins), 400, replace = TRUE)
lambda <- rep(c(2, 8), each = n_bins / 2)
countdata <- data.frame(bin = bin, count = rpois(400, lambda[bin]))
model <- wisp(
  countdata,
  fit_only       = TRUE,
  model.settings = list(LRO_cost = "none"),
  plot.settings  = list(print.plots = FALSE),
  verbose        = FALSE
)
coef(model)
#>     baseline_context_Rt_species_Tns/Blk1 
#>                                 3.890853 
#>     baseline_context_Rt_species_Tns/Blk2 
#>                                 5.061927 
#> baseline_context_tslope_species_Tns/Blk1 
#>                                 7.494465 
#> baseline_context_tpoint_species_Tns/Blk1 
#>                                10.699851