HT_mean: Horvitz-Thompson estimation of the mean

In khabbazian/rdssim: A simple respondent-driven sampling simulator

Description

Horvitz-Thompson estimation of the population mean using the given sequence of the vertices.

Usage

HT_mean(chain, trait.vec, inclusion.probability)

Arguments

chain	a sequence of node indices that represent the referral process.
trait.vec	is a vector of size number of nodes in the network with 0-1 values that for example indicate a specific trait.
inclusion.probability	is a vector of size number of nodes in the network with the corresponding stationary probabilities.

Details

Let x_1, x_2, \cdots, x_n be a sequence of samples from a finite population of size N. Furthermore, let \pi_1, \cdots, \pi_N be the inclusion of probabilities. Then \mu_HT = 1/N*\sum_i=1^n x_i/\pi_x_i.

Value

the Horvitz-Thompson estimation of the population mean.

References

Horvitz, D. G.; Thompson, D. J. (1952) “A generalization of sampling without replacement from a finite universe”, Journal of the American Statistical Association,

khabbazian/rdssim documentation built on May 20, 2019, 9:22 a.m.