Markov chain Monte Carlo without likelihoods.
Proc Natl Acad Sci U S A
Many stochastic simulation approaches for generating observations from a posterior distribution depend on knowing a likelihood function. However, for many complex probability models, such likelihoods are either impossible or computationally prohibitive to obtain. Here we present a Markov chain Monte Carlo method for generating observations from a posterior distribution without the use of likelihoods. It can also be used in frequentist applications, in particular for maximum-likelihood estimation. The approach is illustrated by an example of ancestral inference in population genetics. A number of open problems are highlighted in the discussion.
|Title:||Markov chain Monte Carlo without likelihoods.|
|Keywords:||Algorithms, Biological Evolution, Computer Simulation, DNA, DNA, Mitochondrial, Genetics, Population, Humans, Likelihood Functions, Markov Chains, Models, Biological, Monte Carlo Method, Stochastic Processes|
|UCL classification:||UCL > School of Life and Medical Sciences
UCL > School of Life and Medical Sciences > Faculty of Life Sciences
UCL > School of Life and Medical Sciences > Faculty of Life Sciences > Biosciences (Division of)
UCL > School of Life and Medical Sciences > Faculty of Life Sciences > Biosciences (Division of) > Genetics, Evolution and Environment > UCL Genetics Institute
Archive Staff Only