|제목||[학술세미나] [학과세미나] 11월 30일(금) 11시 학과세미나 안내|
|내용||[학과세미나] 11월 30일(금) 11시 학과세미나 안내
▪제목 : Statistical imputation and its application
▪연사 : 임종호 (연세대 응용통계학과)
▪일시 : 2018년 11월 30일(금) AM 11:00 – 12:00
▪장소 : 25동 405호
Imputation has been popularly used to handle item nonresponse. However, imputation methods can be also applied to solve other statistical problems, for example, statistical matching and two phase sampling. In this presentation, we discuss those applications with repeated imputation methods: multiple and fractional imputation. We first review these two popular imputation methods by focusing on their differences, and then introduce a relatively new fractional imputation, fractional hot deck imputation (FHDI), which is designed for handling multivariate missing data. Afterwards, we will discuss how those imputation methods can be applied to data integration and outcome dependent sampling design. Numerical results from Monte Carlo simulations and real data analysis are presented as illustrative examples.
세미나 안내_181130_임종호.hwp [14KB]