Zelig (version 2.6-4)

setx: Setting Explanatory Variable Values

Description

The setx command uses the variables identified in the formula generated by zelig and sets the values of the explanatory variables to the selected values. Use setx after zelig and before sim to simulate quantities of interest.

Usage

x.out <- setx(object, fn = list(numeric = mean, ordered = median, 
                                others = mode), 
              data = NULL, cond = FALSE, ...)

Arguments

object

the saved output from zelig.

fn

a list of functions to apply to three types of variables: numeric{numeric variables are set to their mean by default, but you may select any mathematical function to apply to numeric variables.} orde

data

a new data frame used to set the values of explanatory variables. If data = NULL (the default), the data frame called in zelig is used.

cond

a logical value indicating whether unconditional (default) or conditional (choose cond = TRUE) prediction should be performed. If you choose cond = TRUE, setx will coerce fn = NULL and ignor

...

user-defined values of specific variables overwriting the default values set by the function fn. For example, adding var1 = mean(data$var1) or x1 = 12 explicitly sets the value of x1 to 12.

Value

• For unconditional prediction, x.out is a model matrix based on the specified values for the explanatory variables. For multiple analyses (i.e., when choosing the by option in zelig, setx returns the selected values calculated over the entire data frame. If you wish to calculate values over just one subset of the data frame, the 5th subset for example, you may use: x.out <- setx(z.out[[5]])

  For conditional prediction, x.out includes the model matrix and the dependent variables. For multiple analyses (when choosing the by option in zelig), setx returns the observed explanatory variables in each subset.

See Also

The full Zelig manual may be accessed online at http://gking.harvard.edu/zelig.

Examples

# Unconditional prediction:
data(turnout)
z.out <- zelig(vote ~ race + educate, model = "logit", data = turnout)
x.out <- setx(z.out)
s.out <- sim(z.out, x = x.out)

# Unconditional prediction with all observations:
x.out <- setx(z.out, fn = NULL)
s.out <- sim(z.out, x = x.out)

# Unconditional prediction with out of sample data:
z.out <- zelig(vote ~ race + educate, model = "logit",
               data = turnout[1:1000,])
x.out <- setx(z.out, data = turnout[1001:2000,])
s.out <- sim(z.out, x = x.out)

# Using a user-defined function in fn:
quants <- function(x)
  quantile(x, 0.25)
x.out <- setx(z.out, fn = list(numeric = quants))

# Conditional prediction:  
library(MatchIt)
data(lalonde)
match.out <- matchit(treat ~ age + educ + black + hispan + married + 
                     nodegree + re74 + re75, data = lalonde)
z.out <- zelig(re78 ~ distance, data = match.data(match.out, "control"), 
               model = "ls")
x.out <- setx(z.out, fn = NULL, data = match.data(match.out, "treat"),
	      cond = TRUE)
s.out <- sim(z.out, x = x.out)