Rejection ABC sampler

class pints.RejectionABC(log_prior)

Implements the rejection ABC algorithm as described in [1].

Here is a high-level description of the algorithm:

\[\begin{split}\begin{align} \theta^* &\sim p(\theta) \\ x &\sim p(x|\theta^*) \\ \textrm{if } s(x) < \textrm{threshold}, \textrm{then} \\ \theta^* \textrm{ is added to list of samples} \\ \end{align}\end{split}\]

In other words, the first two steps sample parameters from the prior distribution \(p(\theta)\) and then sample simulated data from the sampling distribution (conditional on the sampled parameter values), \(p(x|\theta^*)\). In the end, if the error measure between our simulated data and the original data is within the threshold, we add the sampled parameters to the list of samples.

References

[1]

“Approximate Bayesian Computation (ABC) in practice”. Katalin Csillery, Michael G.B. Blum, Oscar E. Gaggiotti, Olivier Francois (2010) Trends in Ecology & Evolution https://doi.org/10.1016/j.tree.2010.04.001

ask(n_samples)

See ABCSampler.ask().

n_hyper_parameters()

Returns the number of hyper-parameters for this method (see TunableMethod).

name()

See pints.ABCSampler.name().

set_hyper_parameters(x)

Sets the hyper-parameters for the method with the given vector of values (see TunableMethod).

Parameters:

x – An array of length n_hyper_parameters used to set the hyper-parameters.

set_threshold(threshold)

Sets threshold error distance that determines if a sample is accepted (if error < threshold).

tell(fx)

See ABCSampler.tell().

threshold()

Returns threshold error distance that determines if a sample is accepted (if error < threshold).