Methoden vergleichen

	Adaptive Weighted Sampling	Adaptives Cluster-Sampling
Fachgebiet	Umfragemethodik	Umfragemethodik
Familie	Process / pipeline	Process / pipeline
Entstehungsjahr	1990s–2000s	1990
Urheber	Building on Thompson (1990) adaptive sampling and classical importance-weighting; adaptive weighting formalised across survey and Monte Carlo literature	Steven K. Thompson
Typ	Probabilistic sampling procedure	Probability-based adaptive sampling design
Wegweisende Quelle	Thompson, S. K. (1990). Adaptive cluster sampling. Journal of the American Statistical Association, 85(412), 1050–1059.	Thompson, S. K. (1990). Adaptive cluster sampling. Journal of the American Statistical Association, 85(412), 1050–1059.
Aliasnamen	AWS, adaptive importance sampling, sequential adaptive weighting, dynamic weighted sampling	ACS, adaptive network sampling, sequential cluster sampling, neighborhood adaptive sampling
Verwandt	6	6
Zusammenfassung	Adaptive weighted sampling is a probabilistic sampling procedure that assigns and iteratively updates inclusion weights for population units based on observed data collected during the sampling process itself. Unlike static weighted sampling — where weights are fixed before data collection from known auxiliary information — adaptive weighting revises probabilities as new information accumulates, concentrating sampling effort on units that contribute most to estimating the target quantity. It is used in survey methodology, simulation studies, and rare-event estimation.	Adaptive cluster sampling (ACS) is a probability-based design in which an initial random sample of units triggers the inclusion of neighboring units whenever a predefined condition — typically a threshold count of a rare attribute — is satisfied. Developed by Steven K. Thompson in 1990, ACS is especially powerful for estimating the abundance or distribution of rare, spatially clustered populations such as endangered species, disease hotspots, or hard-to-reach social groups.