aod (version 1.1-30)

splitbin: Splits Binomial Data into Bernoulli Data

Description

Splits binomial data (n, y) and optional covariates into Bernoulli data (0 or 1) and the corresponding covariates.

Usage

splitbin(formula, data, id = "id")

Arguments

formula

A formula: its left-hand side must be of the form cbind(y, n - y) where the modelled probability is y/n. The right-hand side indicates the covariates which are considered in the ungrouped data set.

data

A data frame where all the variables described in n, y and formula are found.

id

An optional character string naming the identifier (= grouping factor). Default to id.

Value

• A data frame with response and covariates as indicated in the formula, plus the identifier.

Details

Any variable of mode character is converted into a factor before data processing. Each line (with $n > 0$) of the initial data set generates $n$ lines in the final data set with the same covariate pattern, plus an identifier (named id, by default). The value of this identifier is the row name from which it was expanded in the initial data set.

Examples

data(orob1)
  splitbin(cbind(y, n - y) ~ 1, orob1)
  splitbin(cbind(y, n - y) ~ dilution, orob1)