vcdExtra (version 0.9.6)

get_model: Extract Model Formulas from Model Objects or Model Lists

Description

get_model() extracts the model formula or bracket notation from a single loglm or glm object.

Usage

get_model(x, type = c("brackets", "formula"), abbrev = FALSE, ...)
get_models(x, type = c("brackets", "formula"), abbrev = FALSE, ...)

Value

For get_model(): a character string with the model formula or bracket notation.

    For `get_models()`: a named character vector with the model formulas or bracket notations.

Arguments

x

For get_model(): a loglm or glm object. For get_models(): a glmlist or loglmlist object.

type

Type of output: "brackets" for loglinear bracket notation (e.g., "[AB] [C]"), or "formula" for R formula notation. For glm and glmlist objects, only "formula" is meaningful.

abbrev

Logical or integer. If TRUE or a positive integer, abbreviate factor names to that many characters (default 1 when TRUE). Only applies to bracket notation.

...

Additional arguments passed to loglin2string such as sep and collapse.

Details

get_models() does the same for each model in a loglmlist or glmlist object. These are useful for labeling models in summaries and plots.

For loglm objects created by seq_loglm, the bracket notation is stored in the model.string component. For other loglm objects, it is constructed from the margin component using loglin2string.

For glm objects, the formula is extracted using formula().

Model notation

For loglmlist objects created by seq_loglm, the bracket notation distinguishes between models fit to marginal sub-tables and models fit to the full table. Parentheses are used for marginal sub-tables, e.g., "(Class) (Sex)", while square brackets are used for the full table, e.g., "[Class,Sex,Age] [Survived]".

The abbrev argument can be used to abbreviate factor names for more compact display, e.g., "[C,S,A] [S]".

See Also

glmlist, loglmlist, loglin2string, LRstats, seq_mosaic

Other glmlist functions: Kway(), LRstats(), glmlist(), mosaic.glmlist()

Other loglinear models: assoc_graph(), glmlist(), joint(), plot.assoc_graph(), seq_loglm()

Examples

data(Titanic)
tit.joint <- seq_loglm(Titanic, type = "joint")

# Single model
get_model(tit.joint[[2]])
get_model(tit.joint[[4]])
get_model(tit.joint[[4]], abbrev = TRUE)
get_model(tit.joint[[4]], type = "formula")

# Model list
get_models(tit.joint)
get_models(tit.joint, type = "formula")

# With abbreviated factor names
get_models(tit.joint, abbrev = TRUE)
get_models(tit.joint, abbrev = 2)