A Rough Set Approach to Data Imputation and Its Application to a Dissolved Gas Analysis Dataset

Missing values are a common occurrence in a number of real world databases, and various statistical methods have been developed to address this problem, which is referred to as missing data imputation. In the detection and prediction of incipient faults in power transformers using Dissolved Gas Analysis (DGA), the problem of missing values is influential and has resulted in inconclusive decision-making. Previous methods used for handling missing data (e.g., Deleting cases with incomplete information or substituting the missing values with estimated mean scores), although simple to implement, are problematic because those methods may result in biased data models. Fortunately, recent advances in theoretical and computational statistics have led to more feasible techniques to address the missing data problem.