Domain estimation for a censored study variable

Krapavickaitė, Danutė

Use this URL to cite this publication: https://hdl.handle.net/20.500.14911/117325

Domain estimation for a censored study variable

Details

Publication Type (CRIS)

Straipsnis recenzuojamoje užsienio konferencijos medžiagoje / Article in peer-reviewed foreign conference proceedings (P1g)

Publication Type (eLABa)

Straipsnis recenzuotame konferencijos darbų leidinyje / Article published in peer-reviewed conference proceedings (P1d)

VILNIUS TECH Research Priorities and Topics

Fizinių, technologinių ir ekonominių procesų matematiniai modeliai ir metodai / Mathematical models and methods of physical, technological and economic processes (FM03)

Lithuanian Intelligent Specialization

Įtrauki ir kūrybinga visuomenė / Inclusive and creative society (L103)

Author(s)

Title

Domain estimation for a censored study variable

[en]

Danutė Krapavickaitė

Is part of

Proceedings of the 60th ISI World Statistics Congress, 26-31 July 2015, Rio de Janeiro, Brazil

Published In

Year	Start Page	End Page
2015	2422	2425

Publisher

Hague, The Netherlands : International Statistical Institute (ISI), 2015

Extent

p. 2422-2425

URI

URI	Access Rights
https://hdl.handle.net/20.500.14911/117325
Straipsnis	Viso teksto dokumentas (atviroji prieiga) / Full Text Document (Open Access)

Cited references

Science / Art Area

Gamtos mokslai / Natural Sciences (N)

Field of Science / Art

Matematika / Mathematics (N001)

Keywords (en)

Abstract (en)

The estimator for domain totals of a study variable having many zero values is studied in the paper. Design-based model-assisted estimators are usually inefficient in such a situation because of a high variance of a study variable. Therefore, a censored regression model introduced by Tobin is suggested in this paper to be used in the usual domain estimation framework. It is suggested to estimate model parameters using Bayesian inference, and the model-based estimator of domain total is obtained.