mse: Compute the Mean Squared Error of an Estimator

View source: R/mse.R

mseR Documentation

Compute the Mean Squared Error of an Estimator

Description

A generic function to compute the mean squared error of the predicted values under the estimated model. See also rfh for examples.

Usage

mse(object, ...)

## S3 method for class 'fitrfh'
mse(object, type = "pseudo", predType = "reblupbc", B = 100, ...)

Arguments

object

(see methods) an object containing the estimation result, e.g. rfh

...

arguments passed to methods

type

(character) the type of the MSE. Available are 'pseudo' and 'boot'

predType

(character) the type of prediction: c("reblup", "reblupbc")

B

(numeric) number of bootstrap repetitions

Details

Type pseudo is an approximation of the MSE based on a pseudo linearisation approach by Chambers, et. al. (2011). The specifics can be found in Warnholz (2016). Type boot implements a parameteric bootstrap for these methods.

References

Chambers, R., H. Chandra and N. Tzavidis (2011). "On bias-robust mean squared error estimation for pseudo-linear small area estimators". In: Survey Methodology 37 (2), pp. 153–170.

Warnholz, S. (2016): "Small Area Estimaiton Using Robust Extension to Area Level Models". Not published (yet).

Examples

data("grapes", package = "sae")
data("grapesprox", package = "sae")

fitRFH <- rfh(
  grapehect ~ area + workdays - 1,
  data = grapes,
  samplingVar = "var"
)

mseRFH <- mse(fitRFH)
plot(mseRFH)

wahani/saeRobust documentation built on March 16, 2024, 11:28 a.m.