mse_saeOBns: Parametric Bootstrap Mean Squared Error Estimators of Optimum...

In yas-q/msaeOB: Optimum Benchmarking for Multivariate Small Area Estimation

Parametric Bootstrap Mean Squared Error Estimators of Optimum Benchmarking for Univariate Non Sampled Area in Small Area Estimation

Description

Calculates the parametric bootstrap mean squared error estimates of optimum benchmarking for univariate non sampled area in small area estimation

Usage

mse_saeOBns(
  formula,
  vardir,
  weight,
  cluster,
  samevar = FALSE,
  B = 100,
  MAXITER = 100,
  PRECISION = 1e-04,
  data
)

Arguments

formula	an object of class list of formula describe the fitted model
vardir	vector containing sampling variances of direct estimators
weight	vector containing proportion of units in small areas
cluster	vector containing cluster of auxiliary variable
samevar	logical. If TRUE, the varians is same. Default is FALSE
B	number of bootstrap. Default is 1000
MAXITER	maximum number of iterations for Fisher-scoring. Default is 100
PRECISION	coverage tolerance limit for the Fisher Scoring algorithm. Default value is 1e-4
data	dataframe containing the variables named in formula, vardir, and weight

Value

mse.eblup	estimated mean squared errors of the EBLUPs for the small domains based on Prasad Rao
pbmse.eblupOB	parametric bootstrap mean squared error estimates of the optimum benchmark
running.time	time for running function

Examples

## load dataset
data(datamsaeOBns)

# Compute MSE EBLUP and Optimum Benchmark

## Using parameter 'data'
mse_sae = mse_saeOBns(Y1 ~ X1 + X2, v1, w1, c1, data = datamsaeOBns)

## Without parameter 'data'
mse_sae = mse_saeOBns(datamsaeOBns$Y1 ~ datamsaeOBns$X1 + datamsaeOBns$X2,
datamsaeOBns$v1, datamsaeOBns$w1, datamsaeOBns$c1)

## Return
mse_sae$pbmse.eblupOB # to see the MSE Optimum Benchmark estimators

yas-q/msaeOB documentation built on June 23, 2022, 7:10 p.m.