Biological Age Imputation by Data Depth A Proposal and Some Preliminary Results

The biological age is an indicator of the functional condition of an individual's body. Unlike the chronological age, which just measures the time from birth, the biological age of a human is also affected by its medical condition, life habits, some sociodemographic variables, as well as biomarkers. Taking advantage of the statistical concept of depth, which serves as a measurement of the degree of centrality of a multivariate observation with respect to a dataset, we assess the biological age of an individual as the chronological age that would make her selected records as deep as possible when compared with those of other individuals with a chronological age similar to hers. Some direct conclusions of this imputation technique are presented.