Zelig (version 3.5.1)

model.frame.multiple: Extracting the ``environment'' of a model formula

Description

Use model.frame.multiple after parse.par to create a data frame of the unique variables identified in the formula (or list of formulas).

Usage

model.frame.multiple(formula, data, eqn = NULL, ...)

Arguments

formula

a list of formulas of class "multiple", returned from parse.par

data

a data frame containing all the variables used in formula

eqn

an optional character string or vector of character strings specifying the equations (specified in describe.mymodel) for which you would like to pull out the relevant variables.

...

additional arguments passed to model.frame.default

Value

• The output is a data frame (with a terms attribute) containing all the unique explanatory and response variables identified in the list of formulas. By default, missing (NA) values are listwise deleted.

  If as.factor appears on the left-hand side, the response variables will be returned as an indicator (0/1) matrix with columns corresponding to the unique levels in the factor variable. If any formula contains more than one tag statement, model.frame.multiple will return the original variable in the data frame and use the tag information in the terms attribute only.

See Also

model.matrix.default, parse.formula and the full Zelig manual at http://gking.harvard.edu/zelig

Examples

data(sanction)
formulae <- list(import ~ coop + cost + target,
                 export ~ coop + cost + target)
fml <- parse.formula(formulae, model = "bivariate.logit")
D <- model.frame(fml, data = sanction)