fitSTEBLUP: fitSTEBLUP

In wahani/SAE: Small Area Estimation

Description

Wrapper function for the estimation of the spatio-temporal-FH model as in Marhuenda (2012). Uses the function FitSpatioTemporalFH provided by Sample Deliverable 22: Software on Small Area Estimation

Usage

fitSTEBLUP(formula, dat, beta, sigma, rho,
  sigmaSamplingError = seSigmaClosure(nDomains, nTime)(),
  nDomains = getNDomains(dat), nTime = getNTime(dat),
  w0 = w0Matrix(nDomains), w = w0/rowSums(w0), tol = 0.001,
  method = "Nelder-Mead", maxIter = 500)

Arguments

formula	formula for the fixed effects part
dat	data with dependent and independent variable as they are used in formula. The data should contain a variable Domain and Time indicating the domain and time period for each observation respectively
beta,	sigma, rho start values for the parameters
sigmaSamplingError	True variances for the sampling errors
nDomains	the number of domains. Will be determined by default, see dat
nTime	the number of time periods. Will be determined by default, see dat
w0	proximity matrix - not needed in the call to FitSpatioTemporalFH
w	at the moment row-standardized proximity matrix
tol	numerical tolerance for algorithm
method	method used by the function optim - not needed in the call to FitSpatioTemporalFH
maxIter	maximal number of iterations

Examples

require(spatioTemporalData); require(SAE)
set.seed(3)
setup <- simSetupMarhuenda(nDomains=30, nTime=5, sarCorr=0.5, arCorr=0.5, n = 200)
output <- simRunMarhuenda(setup)
dat <- slot(output[[1]], "data")[[1]]
result <- fitSTEBLUP(y~x, dat, c(0,1), c(1,1), c(0.5,0.5))
summary(result)

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