Skip to contents

Summary of an excodeModel

summary-excodeModel-method.Rd

Summarize a fitted excodeModel at time points where excess-count detection was performed, including expectations, posterior probabilities, p-values, and corresponding thresholds per the model specification.

Usage

# S4 method for class 'excodeModel'
summary(
  object,
  pars = c("posterior", "pval", "zscore", "date", "timepoint", "observed", "emission",
    "BIC", "AIC"),
  prob_threshold = 0.5,
  pval_threshold = 0.05,
  anscombe_threshold = 2,
  maxiter = 1000
)

Arguments

object

An excodeModel to summarize.

pars

Character vector of fields to include (e.g., "posterior", "pval", "zscore", "date", "timepoint", "observed", "emission", "BIC", "AIC"); defaults to all.

prob_threshold

Numeric posterior probability threshold used to compute the posterior-based upper bound (posterior_ub) and flag excess counts (>= threshold); default 0.5.

pval_threshold

Numeric p-value threshold used to compute the p-value-based upper bound (pval_ub) and flag excess counts (<= threshold); default 0.01.

anscombe_threshold

Numeric threshold on the Anscombe z-score used for signal assessment; default 2.

maxiter

Integer maximum iterations when estimating the posterior alarm threshold; default 1000.

Value

A data.frame summarizing selected components of the excodeModel (expected values, posterior, p-value, and fit metrics such as AIC/BIC) according to pars.

See also

excodeModel, excodeFamily, excodeFormula

Examples


data(shadar_df)
res_har_pois <- run_excode(surv_ts = shadar_df, timepoints = 295, distribution = "Poisson", 
states = 2, periodic_model = "Harmonic", time_trend = "Linear", set_baseline_state = TRUE)
summary(res_har_pois)
#>   posterior0 posterior1      pval   zscore       date timepoint observed
#> 1  0.8669206  0.1330794 0.1883469 1.141199 2006-08-28       295        6
#>        mu0      mu1      BIC      AIC
#> 1 3.826315 12.43947 1149.244 1066.034