Dynamic nested sampling: an improved algorithm for parameter estimation and evidence calculation

We introduce dynamic nested sampling: a generalisation of the nested sampling algorithm in which the number of “live points” varies to allocate samples more efficiently. In empirical tests the new method significantly improves calculation accuracy compared to standard nested sampling with the same number of samples; this increase in accuracy is equivalent to speeding up the computation by factors of up to $$\sim 72$$∼72 for parameter estimation and $$\sim 7$$∼7 for evidence calculations. We also show that the accuracy of both parameter estimation and evidence calculations can be improved simultaneously. In addition, unlike in standard nested sampling, more accurate results can be obtained by continuing the calculation for longer. Popular standard nested sampling implementations can be easily adapted to perform dynamic nested sampling, and several dynamic nested sampling software packages are now publicly available.

[1]  Hansen.  On the Construction of New Lunar Tables, and on some points in the Lunar Theory depending on the Conformation of the Moon with respect to its Centre of Gravity , 1854 .

[2]  F. Feroz,et al.  Multimodal nested sampling: an efficient and robust alternative to Markov Chain Monte Carlo methods for astronomical data analyses , 2007, 0704.3704.

[3]  Brendon J. Brewer,et al.  Diffusive nested sampling , 2009, Stat. Comput..

[4]  Gautham Narayan,et al.  MOSFiT: Modular Open Source Fitter for Transients , 2017, 1710.02145.

[5]  Daniel Foreman-Mackey,et al.  DNest4: Diffusive Nested Sampling in C++ and Python , 2016, 1606.03757.

[6]  Will Handley,et al.  fgivenx: A Python package for functional posterior plotting , 2018, J. Open Source Softw..

[7]  M. P. Hobson,et al.  polychord: nested sampling for cosmology , 2015, Monthly Notices of the Royal Astronomical Society: Letters.

[8]  Abraham Loeb,et al.  Constraining the stellar mass function from the deficiency of tidal disruption flares in the nuclei of massive galaxies , 2018, Monthly Notices of the Royal Astronomical Society.

[9]  Edward Higson,et al.  Perfectns: Perfect Dynamic and Standard Nested Sampling for Spherically Symmetric Likelihoods and Priors , 2018, J. Open Source Softw..

[10]  Edward Higson,et al.  Bayesian sparse reconstruction: a brute-force approach to astronomical imaging and machine learning , 2018, Monthly Notices of the Royal Astronomical Society.

[11]  A. Savitzky,et al.  Smoothing and Differentiation of Data by Simplified Least Squares Procedures. , 1964 .

[12]  Clément Walter,et al.  Point process-based Monte Carlo estimation , 2014, Stat. Comput..

[13]  Anthony N. Lasenby,et al.  Bayesian model selection without evidences: application to the dark energy equation-of-state , 2015, 1506.09024.

[14]  M. Betancourt.  Nested Sampling with Constrained Hamiltonian Monte Carlo , 2010, 1005.0157.

[15]  Edward Higson,et al.  Sampling Errors in Nested Sampling Parameter Estimation , 2017 .

[16]  Charles R. Keeton,et al.  On statistical uncertainty in nested sampling , 2011, 1102.0996.

[17]  A. Pettitt,et al.  Recursive Pathways to Marginal Likelihood Estimation with Prior-Sensitivity Analysis , 2013, 1301.6450.

[18]  G. W. Pratt,et al.  Planck 2015. XX. Constraints on inflation , 2015, 1502.02114.

[19]  M. P. Hobson,et al.  Importance Nested Sampling and the MultiNest Algorithm , 2013, The Open Journal of Astrophysics.

[20]  J. Skilling.  Nested sampling for general Bayesian computation , 2006 .

[21]  Nial Friel,et al.  Improving power posterior estimation of statistical evidence , 2012, Stat. Comput..

[22]  J. Dunkley,et al.  Comparison of sampling techniques for Bayesian parameter estimation , 2013, 1308.2675.

[23]  Edward Higson,et al.  dyPolyChord: dynamic nested sampling with PolyChord , 2018, J. Open Source Softw..

[24]  C. Robert,et al.  Properties of nested sampling , 2008, 0801.3887.

[25]  Iain Murray.  Advances in Markov chain Monte Carlo methods , 2007 .

[26]  J. Skilling.  Nested Sampling’s Convergence , 2009 .

[27]  B. Yanny,et al.  Dark Energy Survey year 1 results: Cosmological constraints from galaxy clustering and weak lensing , 2017, Physical Review D.

[28]  D. Frenkel,et al.  Superposition Enhanced Nested Sampling , 2014, Physical Review X.

[29]  F. Feroz,et al.  MultiNest: an efficient and robust Bayesian inference tool for cosmology and particle physics , 2008, 0809.3437.

[30]  Edward Higson,et al.  Nestcheck: Error Analysis, Diagnostic Tests and Plots for Nested Sampling Calculations , 2018, J. Open Source Softw..

[31]  Dirk P. Kroese,et al.  Unbiased and consistent nested sampling via sequential Monte Carlo , 2018, 1805.03924.

[32]  Fabian J. Theis,et al.  An adaptive scheduling scheme for calculating Bayes factors with thermodynamic integration using Simpson’s rule , 2015, Statistics and Computing.

[33]  M. Hobson,et al.  nestcheck: diagnostic tests for nested sampling calculations , 2018, Monthly Notices of the Royal Astronomical Society.

[34]  Yuko Okamoto,et al.  Generalized-ensemble algorithms: enhanced sampling techniques for Monte Carlo and molecular dynamics simulations. , 2003, Journal of molecular graphics & modelling.

[35]  A. Lasenby,et al.  polychord: next-generation nested sampling , 2015, 1506.00171.