Model average of derived quantities

Usage

model_average(
  object = NULL,
  weights = NULL,
  uncertainty = FALSE,
  nboot = 10000,
  Rceattle = NULL
)

Arguments

object: list of Rceattle model objects
weights: vector of weights to be used for weighting models
uncertainty: TRUE/FALSE Sample uncertainty across derived quantities using weighted bootstrap from the asymptotic distribution of MLEs
nboot: Number of bootstraps taken from asymptotic distribution of MLEs. Default = 10000
Rceattle: deprecated name for object, still accepted so existing scripts keep working. Supplying both is an error.

Value

an Rceattle object with derived quantities weighted by the specified weights. The length of the derived quantities spans the years which overlap across all models.

Examples

if (FALSE) { # \dontrun{
# Equal weights across two candidate models.
model_average(list(fit1, fit2), weights = c(1, 1))
# Weighted by AIC, with bootstrapped uncertainty.
model_average(list(fit1, fit2), weights = c(0.7, 0.3), uncertainty = TRUE)
} # }