The Census Bureau conducts the monthly Retail Trade Survey of the business universe in order to provide timely estimates of the level and trend sales. The data for each establishment are subjected to a series of edit checks and to be imputed if they are missing. The problem of handling missing data for the Monthly Retail Trade Survey is examined in the paper. The Monthly Retail Trade Survey is composed of a list sample and an area sample, where the list sample contains 96% of the total sample size. The list sample consists of a fixed panels of certainty units (which report every month) and rotating panels of non-certainty sampling units (which report every three months). A stratified random sampling design was used (See (5)). The main variables collected in the rotating panel cases are the monthly retail sales for the current month and the previous month. For fixed panel cases, only current monthly sales are collected. These items are sometimes not reported or suppressed because of edit failure.