Ratio imputation method for handling item-nonresponse in Eichhorn model

Issue title: Statistical Inference with Missing Data

Guest editors: Ma del Mar Rueda

Article type: Research Article

Authors: Zaizai, Yan^{a; b; *} | Jingyu, Wang^{a; c} | Junfeng, Lai^a | Hua, Wanyan^a

Affiliations: [a] Science College of Inner Mongolia university of technology, Hohhot, Inner Mongolia, 010051, P.R. China | [b] Management College of Inner Mongolia university of technology, Hohhot, Inner Mongolia, 010051, P.R. China | [c] Science College of Qiqihaer university, Qiqihaer, Heilongjiang Province, 161006, P.R. China

Correspondence: [*] Corresponding author. E-mail: [email protected].

Abstract: Missing data is often a problem in social science data. Imputation methods fill in the missing responses and lead, under certain conditions, to valid inference. This paper proposes an investigation procedure concerning the Eichhorn and Hayre [6]'s quantitative sensitive characteristic randomized response procedure in the presence of missing data. For the missing randomized responses, their values are imputed in the ratio form. The estimator for the population mean of sensitive characteristic variable is addressed and its nature is discussed in detail. A simulation study evaluates the proposed imputation methods, and this procedure is found to perform well with regards to bias and mean square error.

Keywords: Item-nonresponse, imputation, ratio imputation, randomized response, quantitative sensitive variable, auxiliary variable

DOI: 10.3233/MAS-2008-3202

Journal: Model Assisted Statistics and Applications, vol. 3, no. 2, pp. 89-98, 2008