Markov Dependence in Statistics and Information Theory and Statistical Problems in Physical Mapping
Abstract
Investigations were carried out on Markov dependence in statistics and information theory, and statistical problems in physical mapping. Results were obtained on the Minimum Description Length Principle and statistical inference, adaptive quantization in image compression, Markov chain Monte Carlo methods, and statistical problems in the Human Genome Project. These results shed light on the connections between information theory and statistics, on the role of parametric models in quantization and image compression, on understanding the convergence behaviors of Markov chain Monte Carlo samplers, and on the information needed for a clone map of chromosomes. Furthermore, a wavelet image coder is designed as part of the investigation and it gives an excellent performance on test images.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 31, 1998
- Accession Number
- ADA366138
Entities
People
- Bin Yu
Organizations
- University of California, Berkeley