Zelig (version 3.5.1)

parse.par: Select and reshape parameter vectors

Description

The parse.par function reshapes parameter vectors for comfortability with the output matrix from model.matrix.multiple. Use parse.par to identify sets of parameters; for example, within optimization functions that require vector input, or within qi functions that take matrix input of all parameters as a lump.

Usage

parse.par(par, terms, shape = "matrix", eqn = NULL)

Arguments

par

the vector (or matrix) of parameters

terms

the terms from either model.frame.multiple or model.matrix.multiple

shape

a character string (either "matrix" or "vector") that identifies the type of output structure

eqn

a character string (or strings) that identify the parameters that you would like to subset from the larger par structure

Value

• A matrix or vector of the sub-setted (and reshaped) parameters for the specified parameters given in "eqn". By default, eqn = NULL, such that all systematic components are selected. (Systematic components have ExpVar = TRUE in the appropriate describe.model function.)

  If an ancillary parameter (for which ExpVar = FALSE in describe.model) is specified in eqn, it is always returned as a vector (ignoring shape). (Ancillary parameters are all parameters that have intercept only formulas.)

See Also

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

Examples

# Let's say that the name of the model is "bivariate.probit", and
# the corresponding describe function is describe.bivariate.probit(), 
# which identifies mu1 and mu2 as systematic components, and an 
# ancillary parameter rho, which may be parameterized, but is estimated 
# as a scalar by default.  Let par be the parameter vector (including 
# parameters for rho), formulae a user-specified formula, and mydata
# the user specified data frame.  

# Acceptable combinations of parse.par() and model.matrix() are as follows:
## Setting up
data(sanction)
formulae <- cbind(import, export) ~ coop + cost + target
fml <- parse.formula(formulae, model = "bivariate.probit")
D <- model.frame(fml, data = sanction)
terms <- attr(D, "terms")

## Intuitive option
Beta <- parse.par(par, terms, shape = "vector", eqn = c("mu1", "mu2"))
X <- model.matrix(fml, data = D, shape = "stacked", eqn = c("mu1", "mu2")
eta <- X
## Memory-efficient (compact) option (default)
Beta <- parse.par(par, terms, eqn = c("mu1", "mu2"))    
X <- model.matrix(fml, data = D, eqn = c("mu1", "mu2"))
eta <- X
## Computationally-efficient (array) option
Beta <- parse.par(par, terms, shape = "vector", eqn = c("mu1", "mu2"))
X <- model.matrix(fml, data = D, shape = "array", eqn = c("mu1", "mu2"))
eta <- apply(X, 3, '