Title: Time-varying canonical correlation analysis
Authors: Grace Yoon - Texas A and M University (United States) [presenting]
Irina Gaynanova - Texas A and M University (United States)
Abstract: Canonical correlation analysis (CCA) has been widely used to describe associations between two sets of variables and multiple extensions have been developed for high-dimensional data. However, existing methods cannot be applied to data over time. We present time-varying CCA which can be applied to high-dimensional data and allows us to study how the associations between the two data sets change over time. We also propose data aggregation based on the change point for the small sample size. In addition, the proposed method can take into account variable types of two data sets, such as continuous, zero-inflated and binary. We evaluate the performance of the proposed method in both simulated and real data.