Dealing sensitive characters on successive occasions through a general class of estimators using scrambled response techniques
| Autor: | Pidugu Trisandhya, Kumari Priyanka, Richa Mittal |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
0301 basic medicine
Statistics and Probability Class (set theory) Mean squared error Population mean Privacy protection Estimator 02 engineering and technology Type (model theory) Successive sampling 03 medical and health sciences 030104 developmental biology Character (mathematics) Statistics 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Mathematics |
| Zdroj: | METRON. 76:203-230 |
| ISSN: | 2281-695X 0026-1424 |
| DOI: | 10.1007/s40300-017-0131-1 |
| Popis: | Present article endeavours to propose a general class of estimators to estimate population mean of a sensitive character using non-sensitive auxiliary information under five different scrambled response models in two occasions successive sampling. Various well-known estimators have been modified for the estimation of sensitive population mean and hence they also become a member of proposed general class of estimators. Properties of proposed class of estimators have been derived and checked empirically while comparing the proposed class of estimators with respect to modified Jessen (Iowa Agric Exp Stn Res Bull 304:1–104, 1942) type estimator and modified Singh (Stat Transit 7(1):21–26, 2005) type estimator under five different scrambled response models. The effectiveness of different models has been discussed while comparing it with the direct questioning methods. A model for optimum total cost has also been proposed. Privacy protection has been elaborated for all considered models. Numerical illustrations including simulation studies are abundant to the theoretical results. Finally suitable recommendations are forwarded. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink