What are the advantages over univariate regression?
Univariate regression estimates the association between the target variable and the outcome without accounting for the influence of other variables. This approach often identifies target variables as important that in truth have no impact on the outcome if the dataset contains so-called confounding variables. These are variables that are associated with the target variable as well as with the outcome and that may therefore mask the true relationship between the target variable and the outcome. TargetDiscovery achieves a much lower rate of false discovery by estimating only that part of the association between the target variable and the outcome that cannot be explained by an association between the outcome and confounding variables.