Catalogo Articoli (Spogli Riviste)
OPAC HELP
Titolo: Estimation of the mean from a twodimensional sample: The geostatistical modelbased approach
Autore: Aubry, P; Debouzie, D;
 Indirizzi:
 Univ Lyon 1, CNRS, UMR 5558, F69622 Villeurbanne, France Univ Lyon 1 Villeurbanne France F69622 58, F69622 Villeurbanne, France
 Titolo Testata:
 ECOLOGY
fascicolo: 5,
volume: 82,
anno: 2001,
pagine: 1484  1494
 SICI:
 00129658(200105)82:5<1484:EOTMFA>2.0.ZU;2X
 Fonte:
 ISI
 Lingua:
 ENG
 Soggetto:
 SOIL PROPERTIES; VARIOGRAMS; VARIANCE; UNITS;
 Keywords:
 acorn density; conditional simulation; estimation of the mean by intervals; geostatistics; inferences; designbased; inferences; modelbased; random sampling; simple; random sampling; stratified; spatial autocorrelation; superpopulation model; systematic sampling;
 Tipo documento:
 Article
 Natura:
 Periodico
 Settore Disciplinare:
 Agriculture,Biology & Environmental Sciences
 Citazioni:
 45
 Recensione:
 Indirizzi per estratti:
 Indirizzo: Aubry, P Univ Lyon 1, CNRS, UMR 5558, 43 Bd 11 Novembre 1918, F69622 Villeurbanne,France Univ Lyon 1 43 Bd 11 Novembre 1918 Villeurbanne France F69622 e



 Citazione:
 P. Aubry e D. Debouzie, "Estimation of the mean from a twodimensional sample: The geostatistical modelbased approach", ECOLOGY, 82(5), 2001, pp. 14841494
Abstract
Three common sampling methods for spatial data are simple random sampling (SRS), systematic sampling (Syst), and stratified random sampling (Strat). When values are spatially correlated, SRS usually leads to estimates havinghigher variances than with the other two methods. However, the SRSbased variance can be "designbased," using hypothetical repetitions of the sampling process: it needs no assumptions about the unobserved values. Syst and Strat methods do need "modelbased" assumptions. As a result, variances of estimates based on Syst or Strat are often estimated by the SRS formula, e.g., s(2)/n for the variance of a mean. In a previous paper, we showed that these estimates are misleading, and wedescribed another way to obtain improved estimates. It assumes that the pattern being sampled arose from a random process and considers the expected squared error, E([estimate  parameter](2)), that would arise if this process were run repeatedly and sampled at the same places each time. The unconditional variance uses all such hypothetical runs of the process. This paper extends the previous one to a conditional variance; it uses only those hypothetical runs whose values at the sampled points are the same as those in the actual sample. We show that the conditional variance is morereliable and that the other two overestimate uncertainty, often greatly. We calculate confidence and prediction intervals of the mean from the unconditional and the conditional variances. Intervals based on the conditional variance are usually shorter, and they better describe the true uncertainty. We illustrate the methods by case studies on three populations: one using data on the spatial population of acorns under an isolated oak, and the other two simulated. We emphasize that, although our approach applies to any twodimensional sample, modelbased inference requires that the sample be sufficiently representative of the population and that the probabilistic spatial model be specified with caution.
ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 28/11/20 alle ore 03:53:07