Sliding Window Method for Imputing Missing Data

The rapid technological advancement has led to an increased amount of content and information which needs to be reliably analyzed and correlated in order to make accurate inferences. Data quality is a stringent issue when considering gaps or missing information and various imputation methods, algorithms or statistical techniques have been created to correct missing data. To this purpose we propose an imputation method using the Sliding Window Technique to help infer missing data from datasets, a method that is efficient in terms of accuracy and complexity. In order to illustrate the method, we present a case study that simulates a scenario in which a half-yearly dataset must be analyzed in relation to annual datasets. The obtained results are passed to a statistical descriptive analysis and the conclusions derived from the performed analysis suggest that the estimations are homogeneous enough.