fitfh: Fit Fay-Herriot Model

In wahani/SAE: Small Area Estimation

Description

Fit Fay-Herriot Model

Usage

fitfh(formula, vardir, idName, data, optsRobust = genOptsRobust(),
  optsOptim = genOptsOptim(), type = "RFH", y = TRUE)

Arguments

formula	for fixed effects
vardir	variable name as character of known sampling variances
idName	variable name as character of the domain ID
data	area-level data as data.frame
optsRobust	a list containing parameters for robust estimation. See genOptsRobust for available options
optsOptim	a list containing parameters controlling the algorithm. See genOptsOptim for available options
type	The type of model, see details
y	logical, TRUE if the vector of direct estimates (response) is part of the return value. Necessary for (postestimation) computing residuals etc.

Details

The type controls the model to be estimated. At this time only the robust Fay-Herriot model ('RFH') is supported.

Examples

data(milk)
milk$SD <- milk$SD^2
fit <- fitfh(formula = yi ~ as.factor(MajorArea), vardir="SD", idName="SmallArea", data = milk)

wahani/SAE documentation built on May 3, 2019, 7:37 p.m.