University of Cambridge > > Isaac Newton Institute Seminar Series > Exact Bayesian inference for change point models with application to genomics

Exact Bayesian inference for change point models with application to genomics

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Mustapha Amrani.

Inference for Change-Point and Related Processes

We are interested in the evaluation of uncertainty in the location of change points. Because of the discrete nature of change points, standard statistical theory does not apply easily and only few asymptotic results are available to evaluate quantity of interest such as confidence intervals for change point locations. In a Bayesian framework, such quantities are often obtained via computationally demanding Monte-Carlo techniques. We will present a general Bayesian methodology to compute such quantities in an exact manner, with quadratic complexity. A parallel can be made between this approach and dynamic programming algorithms. This methodology allows to compute a series of posterior distributions, such as this of the total number of breakpoints or this of any change point location. Based on these results we will consider a Bayesian method to compare the location of change points in independent sequences. Eventually, this method will be used to compare the location of transcript boundaries in yeast under different growth condition.

This talk is part of the Isaac Newton Institute Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity