Type: Package
Title: Small Area Estimation with Zero-Inflated Model
Version: 1.0.2
Description: This function produces empirical best linier unbiased predictions (EBLUPs) for Zero-Inflated data and its Relative Standard Error. Small Area Estimation with Zero-Inflated Model (SAE-ZIP) is a model developed for Zero-Inflated data that can lead us to overdispersion situation. To handle this kind of situation, this model is created. The model in this package is based on Small Area Estimation with Zero-Inflated Poisson model proposed by Dian Christien Arisona (2018)https://repository.ipb.ac.id/handle/123456789/92308. For the data sample itself, we use combination method between Roberto Benavent and Domingo Morales (2015)<doi:10.1016/j.csda.2015.07.013> and Sabine Krieg, Harm Jan Boonstra and Marc Smeets (2016)<doi:10.1515/jos-2016-0051>.
License: GPL-3
Encoding: UTF-8
LazyData: true
RoxygenNote: 7.1.1
Imports: stats
Depends: R (≥ 2.10)
URL: https://github.com/dheel/zipsae
BugReports: https://github.com/dheel/zipsae/issues
NeedsCompilation: no
Packaged: 2021-06-14 11:54:21 UTC; User
Author: Fadheel Wisnu Utomo [aut, trl, cre], Ika Yuni Wulansari [aut, ths]
Maintainer: Fadheel Wisnu Utomo <221709671@stis.ac.id>
Repository: CRAN
Date/Publication: 2021-06-14 13:50:02 UTC

Sample Data for Small Area Estimation with Zero-Inflated Poisson model

Description

A Dataset which is generate with Zero-Inflated Poisson method for Small Area Estimation purpose

This data is generated based on Zero-Inflated Poisson with EBLUP based model

Usage

dataSAEZIP

Format

A data frame with 300 rows and 3 variables:

y

Direct Estimation of y

x1

Auxiliary variable of x1

vardir

Sampling Variance of y

EBLUPs under Zero-Inflated Poisson Model

Description

This function produces empirical best linier unbiased predictions (EBLUPs) for Zero-Inflated data and its Relative Standard Error. Small Area Estimation with Zero-Inflated Model (SAE-ZIP) is a model developed for Zero-Inflated data that can lead us to overdispersion situation. To handle this kind of situation, this model is created. The model in this package is based on Small Area Estimation with Zero-Inflated Poisson model proposed by Dian Christien Arisona (2018)<https://repository.ipb.ac.id/handle/123456789/92308>. For the data sample itself, we use combination method between Roberto Benavent and Domingo Morales (2015)<doi:10.1016/j.csda.2015.07.013> and Sabine Krieg, Harm Jan Boonstra and Marc Smeets (2016)<doi:10.1515/jos-2016-0051>.

Usage

zipsae(data, vardir, formula, PRECISION = 1e-04, MAXITER = 100)

Arguments

data

The data frame with vardir, response, and explanatory variables included with Zero-Inflated situation also.

vardir

Sampling variances of direct estimations, if it is included in data frame so it is the vector with the name of sampling variances.if it is not, it is a data frame of sampling variance in order : var1, cov12,.,cov1r,var2,cov23,.,cov2r,.,cov(r-1)(r),var(r)

formula

List of formula that describe the fitted model

PRECISION

Limit of Fisher-scoring convergence tolerance. We set the default in 1e-4

MAXITER

Maximum number of iterations in Fisher-scoring algorithm. We set the default in 100

Value

This function returns a list of the following objects:

estimate

A Vector with a list of EBLUP with Zero-Inflated Poisson model

dispersion

A list containing the following objects:

coefficient

A list containing the following objects:

Examples

##load the dataset in package
data(dataSAEZIP)

##Extract the vardir (sampling error)
dataSAEZIP$vardir -> sError

##Compute the data with SAE ZIP model
formula = (y~x1)
zipsae(data = dataSAEZIP, vardir = sError, formula) -> saezip

saezip$estimate        #to see the result of Small Area Estimation with Zero-Inflated Model
saezip$dispersion$rse  #to see the relative standard error from the estimation
saezip$coefficient$lambda   #to see the estimator which is gained from the non-zero compilation data
saezip$coefficient$omega   #to see the estimator which is gained from the complete compilation data.

head(saezip)