Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
153 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Predicting the statistical error of analog particle tracing Monte Carlo (2404.00315v1)

Published 30 Mar 2024 in physics.comp-ph, cs.NA, and math.NA

Abstract: Large particle systems are often described by high-dimensional (linear) kinetic equations that are simulated using Monte Carlo methods for which the asymptotic convergence rate is independent of the dimensionality. Even though the asymptotic convergence rate is known, predicting the actual value of the statistical error remains a challenging problem. In this paper, we show how the statistical error of an analog particle tracing Monte Carlo method can be calculated (expensive) and predicted a priori (cheap) when estimating quantities of interest (QoI) on a histogram. We consider two types of QoI estimators: point estimators for which each particle provides one independent contribution to the QoI estimates, and analog estimators for which each particle provides multiple correlated contributions to the QoI estimates. The developed statistical error predictors can be applied to other QoI estimators and nonanalog simulation routines as well. The error analysis is based on interpreting the number of particle visits to a histogram bin as the result of a (correlated) binomial experiment. The resulting expressions can be used to optimize (non)analog particle tracing Monte Carlo methods and hybrid simulation methods involving a Monte Carlo component, as well as to select an optimal particle tracing Monte Carlo method from several available options. Additionally, the cheap statistical error predictors can be used to determine a priori the number of particles N that is needed to reach a desired accuracy. We illustrate the theory using a linear kinetic equation describing neutral particles in the plasma edge of a fusion device and show numerical results. The code used to perform the numerical experiments is openly available.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (45)
  1. Code accompanying the paper: ‘Predicting the statistical error of analog particle tracing Monte Carlo’. https://gitlab.kuleuven.be/numa/public/statistical-errors-analog-ptmc. Accessed: 2024-01-23.
  2. Prediction of Statistical Error in Monte Carlo Transport Calculations. Nuclear Science and Engineering, 60(2):131–142, June 1976.
  3. D. Andrieux. Bounding the coarse graining error in hidden Markov dynamics. Applied Mathematics Letters, 25(11):1734–1739, Nov. 2012.
  4. Bayesian inverse problems with Monte Carlo forward models. Inverse Problems and Imaging, 7(1):81–105, 2013.
  5. R. Bellman and G. M. Wing. An introduction to invariant imbedding. Pure and applied mathematics. Wiley, New York, 1975.
  6. Single-ensemble multilevel Monte Carlo for discrete interacting-particle methods. In preparation.
  7. Hybrid Particle-Continuum Numerical Methods for Aerospace Applications, 2011.
  8. Collisional delta-f scheme with evolving background for transport time scale simulations. Phys. Plasmas, 6(12):19, 1999.
  9. R. E. Caflisch. Monte Carlo and quasi-Monte Carlo methods. Acta Numerica, 7:1–49, Jan. 1998.
  10. G. Chen and I. D. Boyd. Statistical Error Analysis for the Direct Simulation Monte Carlo Technique. Journal of Computational Physics, 126(2):434–448, July 1996.
  11. D. R. Cox. The Analysis of Multivariate Binary Data. Applied Statistics, 21(2):113, 1972.
  12. Noise and error analysis and optimization in particle-based kinetic plasma simulations. Journal of Computational Physics, 440:110394, Sept. 2021. arXiv:2102.02377 [physics].
  13. Accuracy and convergence of coupled finite-volume/Monte Carlo codes for plasma edge simulations of nuclear fusion reactors. Journal of Computational Physics, 322:162–182, Oct. 2016.
  14. M. B. Giles. Multilevel Monte Carlo methods. Acta Numerica, 24:259–328, May 2015.
  15. J. Goodman and J. Weare. Ensemble samplers with affine invariance. Communications in Applied Mathematics and Computational Science, 5(1):65–80, Jan. 2010.
  16. C. Graham and D. Talay. Stochastic Simulation and Monte Carlo Methods: Mathematical Foundations of Stochastic Simulation, volume 68. 01 2013.
  17. N. Horsten. Fluid and Hybrid Fluid-Kinetic Models for the Neutral Particles in the Plasma Edge of Nuclear Fusion Devices. Ph.D., KU Leuven, 2019.
  18. R. Indira. Analytical study of leakage estimators in Monte-Carlo simulation of radiation transport. Annals of Nuclear Energy, 15(5):261–269, Jan. 1988.
  19. R. Indira. Optimization of biasing parameters in particle transport simulations. Annals of Nuclear Energy, 16(11):571–576, Jan. 1989.
  20. Cell Escape Probabilities for Markov Processes on a Grid. Submitted.
  21. Statistical error of reactor calculations by the Monte Carlo method. Physics of Atomic Nuclei, 74(14):1871–1877, Dec. 2011.
  22. J. Klotz. Statistical Inference in Bernoulli Trials with Dependence. The Annals of Statistics, 1(2), Mar. 1973.
  23. On the Assessment of Monte Carlo Error in Simulation-Based Statistical Analyses. The American Statistician, 63(2):155–162, May 2009.
  24. Introduction to Monte-Carlo Methods for Transport and Diffusion Equations. Oxford University Press, 2003.
  25. Data Assimilation: A Mathematical Introduction, volume 62 of Texts in Applied Mathematics. Springer International Publishing, Cham, 2015.
  26. Estimation Methods for the Joint Distribution of Repeated Binary Observations. Biometrics, 51(2):562, June 1995.
  27. I. Lux. Systematic Study of Some Standard Variance Reduction Techniques. Nuclear Science and Engineering, 67(3):317–325, Sept. 1978.
  28. I. L. MacDonald. A coarse-grained Markov chain is a hidden Markov model. Physica A: Statistical Mechanics and its Applications, 541:123661, Mar. 2020.
  29. D. B. Macmillan. Comparison of Statistical Estimators for Neutron Monte Carlo Calculations. Nuclear Science and Engineering, 26(3):366–372, Nov. 1966.
  30. B. Mortier. Advanced Monte Carlo simulation and estimation for kinetic neutral particles in the plasma edge of fusion reactors. Ph.D., KU Leuven, 2020.
  31. A study of source term estimators in coupled finite-volume/Monte-Carlo methods with applications to plasma edge simulations in nuclear fusion: analog and collision estimators. Accepted.
  32. Invariant imbedding applied to source term estimation procedures in a simplified 1d0d slab case. Technical report, KU Leuven, 2020.
  33. A review and perspective on a convergence analysis of the direct simulation Monte Carlo and solution verification. Physics of Fluids, 31(6):066101, June 2019.
  34. A Survey of Methods for Analyzing Clustered Binary Response Data. International Statistical Review / Revue Internationale de Statistique, 64(1):89, Apr. 1996.
  35. The EIRENE and B2-EIRENE Codes. Fusion Science and Technology, 47(2):172–186, Feb. 2005.
  36. Time dependent neutral gas transport in tokamak edge plasmas. Journal of Nuclear Materials, 220-222:987–992, Apr. 1995.
  37. Prediction of Statistical Error and Optimization of Biased Monte Carlo Transport Calculations. Nuclear Science and Engineering, 70(3):243–261, June 1979.
  38. D. W. Scott. On optimal and data-based histograms. Biometrika, 66(3):605–610, 1979.
  39. Verification of the history-score moment equations for weight-window variance reduction. Conference INIS-BR-17055, Brazil, 2011.
  40. J. Spanier. An Analytic Approach to Variance Reduction. SIAM Journal on Applied Mathematics, 18(1):172–190, Jan. 1970.
  41. J. Spanier and E. M. Gelbard. Monte Carlo Principles and Neutron Transport Problems. Addison-Wesley Publishing Company, 1969.
  42. Two-phases hybrid model for neutrals. Nuclear Materials and Energy, 18:41–45, Jan. 2019.
  43. U. Wolff. Monte Carlo errors with less errors. Computer Physics Communications, 176(5):383, Mar. 2007. arXiv:hep-lat/0306017.
  44. E. Xekalaki and J. Panaretos. A Binomial Distribution with Dependent Trials and Its Use in Stochastic Model Evaluation. Communications in Statistics - Theory and Methods, 33(12):3043–3058, Jan. 2004.
  45. N. F. Zhang. The Use of Correlated Binomial Distribution in Estimating Error Rates for Firearm Evidence Identification. Journal of Research of the National Institute of Standards and Technology, 124:124026, Oct. 2019.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now