An Application of Multiple-Imputation and Sampling-Based Estimation
Missing data occurs frequently in agricultural household surveys, which can lead to biased and inefficient regression estimates. Multiple-imputation can be used to overcome the missing-data problem. Previous studies applied multiple-imputation to datasets, where only some of the variables have missing observations and the rest of variables have no-missing observations. However, in reality all the variables in a survey might have missing observations. Currently, there is no theoretical or practical guidance to practitioners on how to apply multiple-imputation when all the variables in a data set have missing observations. The objective of this study is to evaluate the impact of alternative multiple-imputation application methods, when all the variables have missing observations. The data for this study is collected through a mail survey of 2,995 farmers in Missouri and Iowa in spring 2011. Two multiple-imputation methods applied in the imputation-step; one with using only the complete observations, the other with using all the observations. The results of the current study show that using all the observations in the imputation-stage, even if they have missingness, produce estimates with lower standard error. Hence, practitioners should use all the observations in the imputation-stage.
When requesting a correction, please mention this item's handle: RePEc:boc:scon12:5. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Christopher F Baum)
If references are entirely missing, you can add them using this form.