1% Generated by roxygen2: do not edit by hand 2% Please edit documentation in R/model-probit-gee.R 3\docType{class} 4\name{Zelig-probit-gee-class} 5\alias{Zelig-probit-gee-class} 6\alias{zprobitgee} 7\title{Generalized Estimating Equation for Probit Regression} 8\arguments{ 9\item{formula}{a symbolic representation of the model to be 10estimated, in the form \code{y ~ x1 + x2}, where \code{y} is the 11dependent variable and \code{x1} and \code{x2} are the explanatory 12variables, and \code{y}, \code{x1}, and \code{x2} are contained in the 13same dataset. (You may include more than two explanatory variables, 14of course.) The \code{+} symbol means ``inclusion'' not 15``addition.'' You may also include interaction terms and main 16effects in the form \code{x1*x2} without computing them in prior 17steps; \code{I(x1*x2)} to include only the interaction term and 18exclude the main effects; and quadratic terms in the form 19\code{I(x1^2)}.} 20 21\item{model}{the name of a statistical model to estimate. 22For a list of other supported models and their documentation see: 23\url{http://docs.zeligproject.org/articles/}.} 24 25\item{data}{the name of a data frame containing the variables 26referenced in the formula or a list of multiply imputed data frames 27each having the same variable names and row numbers (created by 28\code{Amelia} or \code{\link{to_zelig_mi}}).} 29 30\item{...}{additional arguments passed to \code{zelig}, 31relevant for the model to be estimated.} 32 33\item{by}{a factor variable contained in \code{data}. If supplied, 34\code{zelig} will subset 35the data frame based on the levels in the \code{by} variable, and 36estimate a model for each subset. This can save a considerable amount of 37effort. You may also use \code{by} to run models using MatchIt 38subclasses.} 39 40\item{cite}{If is set to 'TRUE' (default), the model citation will be printed 41to the console.} 42 43\item{corstr:character}{string specifying the correlation structure: "independence", 44"exchangeable", "ar1", "unstructured" and "userdefined"} 45 46\item{See}{geeglm in package geepack for other function arguments.} 47 48\item{id:}{where id is a variable which identifies the clusters. The data should be 49sorted by id and should be ordered within each cluster when appropriate} 50 51\item{corstr:}{character string specifying the correlation structure: "independence", 52"exchangeable", "ar1", "unstructured" and "userdefined"} 53 54\item{geeglm:}{See geeglm in package geepack for other function arguments} 55} 56\value{ 57Depending on the class of model selected, \code{zelig} will return 58 an object with elements including \code{coefficients}, \code{residuals}, 59 and \code{formula} which may be summarized using 60 \code{summary(z.out)} or individually extracted using, for example, 61 \code{coef(z.out)}. See 62 \url{http://docs.zeligproject.org/articles/getters.html} for a list of 63 functions to extract model components. You can also extract whole fitted 64 model objects using \code{\link{from_zelig_model}}. 65} 66\description{ 67Generalized Estimating Equation for Probit Regression 68} 69\details{ 70Additional parameters avaialable to this model include: 71\itemize{ 72 \item \code{weights}: vector of weight values or a name of a variable in the dataset 73 by which to weight the model. For more information see: 74 \url{http://docs.zeligproject.org/articles/weights.html}. 75 \item \code{bootstrap}: logical or numeric. If \code{FALSE} don't use bootstraps to 76 robustly estimate uncertainty around model parameters due to sampling error. 77 If an integer is supplied, the number of boostraps to run. 78 For more information see: 79 \url{http://docs.zeligproject.org/articles/bootstraps.html}. 80} 81} 82 83\examples{ 84data(turnout) 85turnout$cluster <- rep(c(1:200), 10) 86sorted.turnout <- turnout[order(turnout$cluster),] 87z.out1 <- zelig(vote ~ race + educate, model = "probit.gee", 88id = "cluster", data = sorted.turnout) 89summary(z.out1) 90 91} 92\seealso{ 93Vignette: \url{http://docs.zeligproject.org/articles/zelig_probitgee.html} 94} 95