Speaker
Description
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.
