On sampling with desired inclusion probabilities of first and second order

• We present a new simple approximation of target probabilities pi for conditional Poisson sampling to obtain given inclusion probabilities. This approximation is based on the fact that the Sampford design gives inclusion probabilities as desired. Some alternative routines to calculate exact pi-values are presented and compared numerically. Further we derive two methods for achieving prescribed 2nd order inclusion probabilities. First we use a probability function belonging to the exponential family. The parameters of this probability function are determined by using an iterative proportional fitting algorithm. Then we modify the conditional Poisson probability function with an additional quadratic factor.

Subject headings

NATURVETENSKAP  -- Matematik -- Sannolikhetsteori och statistik (hsv//swe)

NATURAL SCIENCES  -- Mathematics -- Probability Theory and Statistics (hsv//eng)

Keyword

Mathematical statistics

Matematisk statistik

Mathematical Statistics

matematisk statistik

Publication and Content Type

vet (subject category)

rap (subject category)