Ridge Fusion in Statistical Learning
This work addresses the need for improved precision matrix estimation in statistical learning applications like clustering and classification, but it appears incremental as it builds on existing penalized methods.
The authors tackled the problem of jointly estimating multiple precision matrices for quadratic discriminant analysis and model-based clustering by proposing a penalized likelihood method with ridge and ridge fusion penalties, resulting in a method that introduces shrinkage and promotes similarity between estimates, though no concrete numbers are provided.
We propose a penalized likelihood method to jointly estimate multiple precision matrices for use in quadratic discriminant analysis and model based clustering. A ridge penalty and a ridge fusion penalty are used to introduce shrinkage and promote similarity between precision matrix estimates. Block-wise coordinate descent is used for optimization, and validation likelihood is used for tuning parameter selection. Our method is applied in quadratic discriminant analysis and semi-supervised model based clustering.