sondage: Survey Sampling Algorithms

Fast implementations of survey sampling algorithms for drawing samples from finite populations. All functions return indices for easy subsetting.

Equal Probability Sampling

• srs() - Simple random sampling (with/without replacement)

• systematic() - Systematic sampling

• bernoulli() - Bernoulli sampling (random size)

Unequal Probability Sampling

Functions prefixed with up_ take either inclusion probabilities (pik, values in [0,1] summing to n) or raw size measures (x, non-negative values):

pik interface (inclusion probabilities):

• up_maxent() - Maximum entropy / Conditional Poisson Sampling

• up_brewer() - Brewer's method

• up_systematic() - Systematic PPS

• up_poisson() - Poisson sampling (random size)

x interface (raw size measures):

• up_multinomial() - PPS with replacement

• up_chromy() - PPS with minimum replacement

Use inclusion_prob() to convert size measures to inclusion probabilities.

Utilities

• inclusion_prob() - Compute inclusion probabilities from size measure

Joint Inclusion Probabilities

• up_maxent_jip() - Exact CPS joint probabilities

• up_brewer_jip() - Brewer & Donadio approximation

• up_systematic_jip() - Exact systematic joint probabilities

• up_poisson_jip() - Independent selections

• up_chromy_pairexp() - Pairwise expectations for Probability Minimum Replacement

References

Tillé, Y. (2006). Sampling Algorithms. Springer Series in Statistics.

See also

Useful links:

• https://dickoa.gitlab.io/sondage

• https://gitlab.com/dickoa/sondage

• Report bugs at https://gitlab.com/dickoa/sondage/-/issues

Author

Maintainer: Ahmadou Dicko mail@ahmadoudicko.com (ORCID)