The interactions of silverfish with their host ants were evaluated in a multivariate approach including all observed behaviours.
Confounding was assessed by a multivariate logistic regression approach; all variables with a p-value ≤0.15 in the univariate analysis were included in the multivariate model and retained in the final model according to a log-likelihood-ratio test for goodness-of-fit.
Our method differs from [ 9] in the following two aspects: Although both works adopt multivariate approach, first of all, in their information-based approach, clustering and cluster selection are done simultaneously, resulting in a set of clusters optimizing the Minimum Description Length.
As there are various imbalance indicators, the optimal approach would be to develop a multivariate approach to jointly incorporate all the informational content of the indicators.
We then analyzed all phenotypes using a multivariate approach (see figure 4).
All statistical models considered in the multivariate approach will be formulated as logistic regression models, with CHD as response, and the symptoms and signs as explanatory variables.
The data availability from a sufficiently large number of studies allows to apply this multivariate approach, which requires reported estimates for the bivariate relations among all proxy variables.
Using a multivariate approach has the advantage of being able to describe how all traits interact to determine fitness.
All these techniques are strongly supported by a multivariate approach.
Moreover, in situations where the number of traits is large [ e.g., expression QTL (eQTL) analysis], it may not be possible to include all the traits in one analysis, using a usual multivariate approach.
