var_pois function

Variance estimator for a Poisson sampling design

var_pois estimates the variance of the estimator of a total for a Poisson sampling design.

var_pois(y, pik, w = rep(1, length(pik)))

Arguments

• y: A (sparse) numerical matrix of the variable(s) whose variance of their total is to be estimated.
• pik: A numerical vector of first-order inclusion probabilities.
• w: An optional numerical vector of row weights (see Details).

Returns

The estimated variances as a numerical vector of size the number of columns of y.

Details

w is a row weight used at the final summation step. It is useful when var_pois is used on the second stage of a two-stage sampling design applying the Rao (1975) formula.

References

Rao, J.N.K (1975), "Unbiased variance estimation for multistage designs", Sankhya, C n°37

Author(s)

Martin Chevalier