Missing Value Estimation for High-Dimensional Data

• Activity Number: 246 - Data Science
• Type: Contributed
• Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Section on Statistical Computing
• Author(s): Mian Arif Shams Adnan and Silvia Irin Sharna*
• Companies: Bowling Green State University and Bowling Green State University
• Keywords: Likelihood Function

Abstract

Since a high dimensional missing value resembles not only an unknown high dimensional data of an unknown high dimensional probability distribution but also their unknown characteristics, it is better to construct a basket of characteristics based on assumed high dimensional missing values. The missing technique, as demonstrated by Sharna et al (2016), is a kind of check and balance method for estimating a missing value. In this paper we offer an extended version of the iterative estimation method for high dimensional missing value. This paper also demonstrates a resampling method for generating 1 or 2 correlated observations from the same high dimensional distribution from where the original sample is drawn.