Archive for demarginalisation

Bangalore workshop [ಬೆಂಗಳೂರು ಕಾರ್ಯಾಗಾರ]

Posted in pictures, R, Running, Statistics, Travel, University life, Wines with tags , , , , , , on July 31, 2014 by xi'an

mathdeptSecond day at the Indo-French Centre for Applied Mathematics and the workshop. Maybe not the most exciting day in terms of talks (as I missed the first two plenary sessions by (a) oversleeping and (b) running across the campus!). However I had a neat talk with another conference participant that led to [what I think are] interesting questions… (And a very good meal in a local restaurant as the guest house had not booked me for dinner!)

To wit: given a target like

\lambda \exp(-\lambda) \prod_{i=1}^n \dfrac{1-\exp(-\lambda y_i)}{\lambda}\quad (*)

the simulation of λ can be demarginalised into the simulation of

\pi (\lambda,\mathbf{z})\propto \lambda \exp(-\lambda) \prod_{i=1}^n \exp(-\lambda z_i) \mathbb{I}(z_i\le y_i)

where z is a latent (and artificial) variable. This means a Gibbs sampler simulating λ given z and z given λ can produce an outcome from the target (*). Interestingly, another completion is to consider that the zi‘s are U(0,yi) and to see the quantity

\pi(\lambda,\mathbf{z}) \propto \lambda \exp(-\lambda) \prod_{i=1}^n \exp(-\lambda z_i) \mathbb{I}(z_i\le y_i)

as an unbiased estimator of the target. What’s quite intriguing is that the quantity remains the same but with different motivations: (a) demarginalisation versus unbiasedness and (b) zi ∼ Exp(λ) versus zi ∼ U(0,yi). The stationary is the same, as shown by the graph below, the core distributions are [formally] the same, … but the reasoning deeply differs.

twoversions

Obviously, since unbiased estimators of the likelihood can be justified by auxiliary variable arguments, this is not in fine a big surprise. Still, I had not thought of the analogy between demarginalisation and unbiased likelihood estimation previously. Continue reading

Follow

Get every new post delivered to your Inbox.

Join 669 other followers