RCRatio {CREDS} | R Documentation |
Population ratio estimator under two-phase random sampling design has gained enormous popularity in present era. This package provides functions for estimation calibrated population ratio under two phase sampling design, including the approximate variance of the ratio estimator. The improved ratio estimator can be applicable for both the case, when auxiliary data is available at unit level or aggregate level for first phase sampled. Calibration weight of each unit of the second phase sample was calculated. Single and combined inclusion probabilities were also estimated for both phases under two phase random sampling. The improved ratio estimator's percentage coefficient of variation was also determined as a measure of accuracy.
RCRatio(N, FSU, SSU)
N |
Population size |
FSU |
First stage sampling units |
SSU |
Second stage sampling units |
CalEstimate: Estimate value of calibration estimator
CalVariance: Variance of calibration estimator
CV: Coefficient of variance
SampleSize: Sample Size of FSU and SSU
DesignWeight: Design weight vector
InclusionProb: Inclusion probability vector
Correlation: Correlation value
Islam, S., Chandra, H., Sud, U.C., Basak, P., Ghosh, N. and Ojasvi, P.R. (2021). A Revised Calibration Weight based Ratio Estimator in Two-phase Sampling: A Case when Unit Level Auxiliary Information is Available for the First-phase Sample, Journal of Indian Society of Agricultural Statistics, 75(2), 147–156.
Ozgul, N. (2021). New improved calibration estimator based on two auxiliary variables in stratified two-phase sampling. Journal of Statistical Computation and Simulation, 91(6), 1243-1256.
f1<-rnorm(100,20,5)
f2<-rnorm(100,20,5)
fsu<-cbind(f1,f2)
s1<-rnorm(50,20,5)
s2<-rnorm(50,20,5)
s3<-rnorm(50,20,5)
s4<-rnorm(50,20,5)
ssu<-cbind(s1,s2,s3,s4)
RCRatio(N=1000, FSU=fsu, SSU=ssu)