Date: Wednesday, April 18th, 3 PM

Location: Room 49 Hardin Hall North

Title: Comparison of Hot deck and Multiple Imputation based on Latent Class Model for Categorical Data

Presented by: Yiyue Xu

ABSTRACT:

Missing value or incomplete data is a common problem in survey research, as nonresponse. In statistics, imputation is the substitution of some value for missing data instead of discarding them. Once all missing values have been imputed, the dataset can then be analyzed using standard techniques for complete data. The purpose of this project was to compare the two different imputation methods, which are Hot-deck Imputation and Multiple Imputation based on Latent Class Model applying on social survey. One is widely used by the government and the other is a new method proposed in 2008.