Zero-and-one inflated binomial family

Zero-and-one inflated binomial for proportions with excess zeros AND ones. Commonly needed in ecological applications where both absence (structural) and perfect detection (saturation) have distinct processes.

Usage

ratiod_zoibinomial(link_num = "logit", link_zi = "logit", link_oi = "logit")

Arguments

link_num

Link function for success probability (default: "logit")

link_zi

Link function for zero-inflation probability (default: "logit")

link_oi

Link function for one-inflation probability (default: "logit")

Value

A ratiod_family object with zero-and-one inflation

Details

The ZOIB (Zero-and-One Inflated Binomial) model has three components:

1. Zero process: P(structural zero) = pi_0

2. One process: P(structural one | not zero) = pi_1

3. Binomial process: P(Y=y | non-structural) follows binomial

Use when both boundaries (0 and n) have excess observations.

Examples

# Create family object
fam <- ratiod_zoibinomial()
print(fam)
#> tulpaRatio family: zoibinomial 
#> [Zero-inflated model]
#> Zero-and-one inflated binomial, fixed trials denominator 
#> 
#> Numerator:  binomial(logit)
#> ZI prob:    bernoulli(logit)
#> Denominator: fixed (fixed)