Brendon J. Brewer

icon indicating a website link https://brendonbrewer.com/

icon company The University of Auckland icon location Auckland, New Zealand icon location Senior lecturer in statistics at the University of Auckland, New Zealand. I work on Bayesian Inference and related topics.

Results 24 comments of Brendon J. Brewer

1d Gauss example -- returns log(Z)=-0.32, instead of 0?

The return value of 'perturb' is used to implement non-uniform priors. For example, here is an example with a Normal(0, 1) prior. The move I use implies a uniform prior,...

1d Gauss example -- returns log(Z)=-0.32, instead of 0?

Hi Johannes, The gaussian in 10D might make Diffusive Nested Sampling look bad, because it's one where alternative methods work really well. ;-) In low dimensions you won't need so...

1d Gauss example -- returns log(Z)=-0.32, instead of 0?

Thanks for PyDNest. That looks very useful! Many people would be confident implementing their model in Python but not C++. I wonder if it will work with the multithreading, though.

1d Gauss example -- returns log(Z)=-0.32, instead of 0?

I am impressed! The ideas aren't that complicated, but whenever I've tried to implement it in another language I gave up because it became too complex. I'll let you know...