1. Sparse semiparametric canonical correlation analysis for data of mixed types
- Author
-
Grace Yoon, Raymond J. Carroll, and Irina Gaynanova
- Subjects
FOS: Computer and information sciences ,Statistics and Probability ,Rank (linear algebra) ,General Mathematics ,01 natural sciences ,Data type ,Article ,Methodology (stat.ME) ,010104 statistics & probability ,03 medical and health sciences ,Bayesian information criterion ,Applied mathematics ,0101 mathematics ,Statistics - Methodology ,030304 developmental biology ,Mathematics ,Parametric statistics ,0303 health sciences ,Covariance matrix ,Applied Mathematics ,Estimator ,Agricultural and Biological Sciences (miscellaneous) ,Transformation (function) ,Statistics, Probability and Uncertainty ,General Agricultural and Biological Sciences ,Canonical correlation - Abstract
Canonical correlation analysis investigates linear relationships between two sets of variables, but often works poorly on modern data sets due to high-dimensionality and mixed data types such as continuous, binary and zero-inflated. To overcome these challenges, we propose a semiparametric approach for sparse canonical correlation analysis based on Gaussian copula. Our main contribution is a truncated latent Gaussian copula model for data with excess zeros, which allows us to derive a rank-based estimator of the latent correlation matrix for mixed variable types without the estimation of marginal transformation functions. The resulting canonical correlation analysis method works well in high-dimensional settings as demonstrated via numerical studies, as well as in application to the analysis of association between gene expression and micro RNA data of breast cancer patients., Accepted to Biometrika. Main text: 19 pages and 3 figures. Supplementary material: 28 pages and 9 figures
- Published
- 2020