Neatbildētības novirzes analīze izlases veida apsekojumos
Loading...
Date
Authors
Advisor
Journal Title
Journal ISSN
Volume Title
Publisher
Latvijas Universitāte
Language
N/A
Abstract
Tika izvēlēta diplomdarba tēma "Neatbildētības novirzes analīze izlases veida apsekojumos", jo neatbildētības problēma ir īpaši aktuālā statistikajos pētījumos. Darba uzdevums bija aplūkot neatbildētības teoriju un izpētīt palīginformīcijas izmantošanas metodi, kas ļauj samazināt novirzi. Darba mērķis bija, veicot eksperimentus ar dažādiem palīgvektoriem, salīdzināt kurš no tiem dod vislabāku nerespondences novirzes korekciju, izmantojot trīs faktorus. Visi aprēķini tika veikti brīvpieejas programmā R.
Atslēgas vārdi: Horvitza-Tompsona novērtējums, GREG novērtējums, Ratio novērtējums, neatbildētība, novirzes indikators, palīginformācija.
The topic of diploma thesis was chosen "Nonresponse bias analysis in survey sampling", because nonresponse problem is particularly acute in statistical surveys. Main task was to look at nonresponse theory and survey methods, which allows to reduce bias. The goal was, through experiments with different auxiliary vectors compare which of them gives the best nonresponse adjustment using three factors. All calculations were performed in programm R. Keywords: Horvitz-Thompson estimator, GREG estimator, Ratio estimator, nonresponse, bias indicator, auxiliary information.
The topic of diploma thesis was chosen "Nonresponse bias analysis in survey sampling", because nonresponse problem is particularly acute in statistical surveys. Main task was to look at nonresponse theory and survey methods, which allows to reduce bias. The goal was, through experiments with different auxiliary vectors compare which of them gives the best nonresponse adjustment using three factors. All calculations were performed in programm R. Keywords: Horvitz-Thompson estimator, GREG estimator, Ratio estimator, nonresponse, bias indicator, auxiliary information.