|
|
 |
| |
|
|
|
On the Use of Compromised Imputation for Missing data using Factor-Type Estimators |
|
|
|
PP: 105-113 |
|
|
|
Author(s) |
|
|
|
Priyanka Singh,
Ajeet Kumar Singh,
V. K. Singh,
|
|
|
|
Abstract |
|
|
| There are several methods of handling missing data in sample surveys, which is a typical problem of non-response.
Imputation (fill-in) method is one of the methods to deal with non-response. The term Imputation refers to the process of assigning
one or more values to an item when there is no reported value for that item. Many forms of imputation are available, including mean
imputation, ratio method of imputation, hot deck imputation, cold deck imputation, regression imputation, etc. In recent past, a
number of efficient compromised imputation strategies has been proposed by several survey statisticians.This paper suggests a
one-parameter family of estimators, popularly known as Factor-Type Estimator (FTE), with compromised imputation strategy and
discusses its properties. The proposed strategy has been observed to be more precise than other compromised estimators under
optimality conditions. To support the discussed results, the relative efficiencies of the estimator have been obtained using four sets of
empirical data. |
|
|
|
|
|
|
|
