checkcalibration: Check calibration

In sampling: Survey Sampling

Check calibration

Description

Checks the validity of the calibration. In some cases, the computed g-weights do not allow calibration and the calibration estimators do not exist.

Usage

checkcalibration(Xs, d, total, g, EPS=1e-6)

Arguments

Xs	matrix of calibration variables.
d	vector of initial weights.
total	vector of population totals.
g	vector of g-weights.
EPS	control value used to check the calibration, by default equal to 1e-6.

Details

In the case where calibration is not possible, the 'value' indicates the difference in obtaining the calibration.

Value

The function returns the following three objects:

message	a message concerning the calibration,
result	TRUE if the calibration is possible and FALSE, otherwise.
value	value of max(abs(tr-total)/total, which is used as criterium to validate the calibration, where tr=crossprod(Xs, g*d). If the total vector contains zeros, the value is max(abs(tr-total)).

See Also

calib

Examples

# matrix of auxiliary variables
Xs=cbind(c(1,1,1,1,1,0,0,0,0,0),c(0,0,0,0,0,1,1,1,1,1),c(1,2,3,4,5,6,7,8,9,10))
# inclusion probabilities
pik=rep(0.2,times=10)
# vector of totals
total=c(24,26,280)
# g-weights
g=calib(Xs,d=1/pik,total,method="raking")
# check if the calibration is possible
checkcalibration(Xs,d=1/pik,total,g)

sampling documentation built on Nov. 2, 2023, 6:26 p.m.