Multivariate imputation for missing data handling a case study on small and large data sets

Abstract

Abscent of records generally termed as missing data which should be treated properly before analysis procedes in data analysis. There were many researchers who undoubtedly mislead their research findings without proper treatment of missing data, therefore this review research try to explain the best ways of missing data handling using r programming. Generally, many researchers apply mean and median imputation but this sometimes creates bios in many  situations, therefore, the researcher tries to explain some basic  association among other research variables with treating missing data using r programming. The imputation process suggests five alternatives be replaced for missing data values were generated automatically and substituted easily  at the process of data cleaning and data preparation. Here researcher explains two sample data for missing treatment  and explains many ways for  graphical interpretation  of them. The first data set with 12 observation describes the easiest way of missing replacement and the second  vehicle failure data from internet of 1624 records, whose missing pattern were calculated and replaced with to the respective data sets before analysis.