Performing Horovitz-Thompson Estimation in Spatial Sampling : a Computer-intensive Perspective

• Autorzy: Lorenzo Fattorini , Caterina Pisani
• Języki publikacji: EN

Abstrakty

EN

Spatial sampling is widely applied in economic and social surveys as when, for example, population units are administrative districts in a study region, or farms in an agricultural country or areas in a city. The present chapter proposes the use of some sequential sampling schemes which may be suitable to handle the main problems of spatial sampling, i.e. the presence of spatial correlation as well as the presence of units of unequal sizes. Unfortunately, these proposals involve algebraic complications in the expressions of inclusion probabilities. Hence, the Horvitz-Thompson estimator is generally inapplicable. However, when dealing with with-frame populations and when the inclusion probabilities do not depend on the values of the interest variable in the population, the problem may be straightforwardly bypassed. Indeed, in these cases, the sample selection can be independently replicated for an adequate number of times, in such a way that the population values are quantified in the field only for the units in the first sample, while the remaining samples are used to estimate the first- and second-order inclusion probabilities on the basis of the proportion of times in which the units or the couples of units enter the selected samples. (fragment of text)

Artykuł dotyczy wykorzystania statystyki Horovitza-Thompsona do estymacji przeciętnej na podstawie próby losowanej przestrzennie w perspektywie intensywnego wykorzystania komputerów.

Słowa kluczowe

EN

Sampling survey; Estimation; Spatial analysis

PL

Badania reprezentacyjne; Estymacja; Analiza przestrzenna

• Czasopismo: Prace Naukowe / Akademia Ekonomiczna w Katowicach
• Rocznik: 2004
• Tom: Metoda reprezentacyjna w badaniach ekonomiczno-społecznych
• Strony: 19--31

Twórcy

autor

Lorenzo Fattorini

• Universita di Siena, Italy

autor

Caterina Pisani

• Universita di Siena, Italy

Bibliografia

• Barabesi L., Fattorini L., Ridolfi G. (1997). Two-phase surveys of elusive populations. In Proceedings of the Statistics Canada Symposium 97: New Direction in Surveys and Censuses, p. 285-287. Statistics Canada, Ottawa.
• Brewer K.R.W., Hanif M. (1983). Sampling with unequal probabilities. Springer-Verlag, New York.
• Bryant D., Chang Y, Rodger C.A., Wei R. (2002). Two-dimensional balanced sampling plans excluding contiguous units. Commun. Statist.-Theory Meth, 31, 1441-1455.
• Cochran W.G. (1977). Sampling Techniques. Wiley, New York.
• Fattorini L., Ridolfi G. (1997). A sampling design for areal units based on spatial variability. Metron, 55, 59-72.
• Fattorini L. (2004). Performing Horvitz-Thompson criterion in complex design: a computer intensive perspective for estimating inclusion probabilities. Submitted. Biometrika.
• Hedayat A.S., Rao C.R., Stufken J. (1988). Sampling plans excluding contiguous units. J. Statist. Plann. Infer., 19, 159-170.
• Hedayat A.S., Sinha B.K. (1991). Design and inference in finite population sampling. Wiley, New York.
• Skalski J.R. (1994). Estimating wildlife populations based on incomplete area surveys. Wildl. Soc. Bull., 22, 192-203.
• Wywiał J. (1996). On space sampling. Statistics in Transition, 2, 1185-1191.
• Wolter K.M. (1985). Introduction to variance estimation. Springer-Verlag, New York.