Update Probability Mass Function with Likelihood Surface

Updates a spatial probability mass function (PMF) using a likelihood surface following Bayesian principles. The function combines: - Prior: Initial spatial PMF representing prior beliefs - Likelihood: Surface showing relative intensity of observations - Posterior: Updated PMF incorporating both prior and likelihood

Values in the posterior will be higher where both prior probability and likelihood are high, representing areas supported by both prior beliefs and observed data.

Usage

update_pmf(prior, likelihood, normalize = TRUE, density = FALSE, snap = FALSE)

Arguments

prior

SpatRaster containing prior PMF

likelihood

SpatRaster containing likelihood surface

normalize

Logical; if TRUE ensure output sums to 1 (default TRUE)

density

Logical; if TRUE return density surface instead of PMF

snap

Logical; if TRUE skip validation

Value

SpatRaster containing either: - Posterior PMF if density = FALSE - Posterior density if density = TRUE

Examples

if (FALSE) { # \dontrun{
# Create prior from population
prior <- transform_pmf(pop)

# Create likelihood from cases
likelihood <- transform_likelihood(
  case_spatial,
  value_col = "cases",
  template = pop
)

# Update prior with likelihood
posterior <- update_pmf(prior, likelihood)

# Get density surface
density <- update_pmf(prior, likelihood, density = TRUE)
} # }