mphase.gain {forestinventory}

mphase.gain

Description

mphase.gain takes as input an object created by the estTable function and returns a validation of which multiphase method and estimator performed best in comparison to the onephase estimation (baseline) in terms of estimation precision.

Usage

mphase.gain(esttable.obj, pref.vartype = "g_variance", exclude.synth = TRUE)

Arguments

esttable.obj

an object of class esttable created by the estTable function

pref.vartype

preferred type of multiphase variance that should be compared to the onephase variance, if more then one variance type has been calculated in the multiphase estimation object(s) stored in esttable. Valid input values are "g_variance" (default) and "ext_variance".

exclude.synth

logical. If set to TRUE (default), synthetic estimations are not considered in the validation.

Value

mphase.gain returns a data.frame containing the following components:

• area: in case of small area estimation: the name of the small area

• var_onephase: standard error of the onephase estimation

• var_multiphase: smallest variance among the (set of) multiphase estimations stored in esttable.obj

• estimator: multiphase estimator with the smallest variance

• method: estimation Method of the multiphase estimator with the smallest variance

• gain: the gain is the reduction (if value is positive) or possibly also the increase (if value is negative) in variance when applying the multiphase as alternative to the onephase estimation

• rel.eff: the relative efficiency defined as the ratio between the onephase variance and the multiphase variance

Note

The gain can be interpreted as: "The multiphase estimation procedure leads to a gain % reduction in variance compared to the onephase procedure".

The relative efficiency can be interpreted as: "Using the onephase estimation procedure, the terrestrial sample size would have to be rel.eff times larger in order to achieve the same precision (in terms of variance) as the mutiphase estimation procedure".