A comparison of a generic MCMC-based algorithm for Bayesian estimation in C++, R and Julia: application to plant growth modeling
Abstract
Plant growth is understood through the use of dynamical systems involving many interacting processes and model parameters whose estimation is therefore a crucial issue, all the more so since experimental data obtained from agronomical systems are most of the time characterized by their scarcity and their heterogeneity owing to the complex underlying acquisition. One of the approaches used to solve this kind of problem within a Bayesian paradigm involves Markov Chain Monte Carlo (MCMC) algorithms, one of the drawbacks of the latter being that they lead to some intensive computation because of the statistical framework employed, which is why the efficiency of the implemented computing methods is of particular importance. In this paper, we compare three implementations of a generic MCMC-based algorithm for Bayesian estimation in C++, R and Julia so as to compare the performance and precision of these languages. Here, genericness means that the estimation algorithm can be used for any dynamic model provided that it is implemented in a given modeling template. Such genericness is of crucial importance in a scientific field such as plant growth modeling for which no reference model exists and new models are constantly developed and evaluated. The tests are conducted for the particular cases of Lotka--Volterra model and the Log-Normal Allocation and Senescence model for sugar beet.
References
[1]
C++ website. http://www.cplusplus.com/.
[2]
Julia website. http://julialang.org/.
[3]
R website. http://www.r-project.org/.
[4]
Andrieu, C., Doucet, A., and Holenstein, R. Particle Markov chain Monte Carlo methods. Journal of the Royal Statistical Society 72 (2010), 269--342.
[5]
Berg, B. A., and Billoire, A. Markov chain monte carlo simulations. In Wiley Encyclopedia of Computer Science and Engineering. John Wiley and Sons, Inc., 2007.
[6]
Bezanson, J., Edelman, A., Karpinski, S., and Shah, V. Julia: A fresh approach to numerical computing. arXiv:1411.1607v1.
[7]
Bezanson, J., Karpinski, S., Shah, V., and Edelman, A. Julia: A fast dynamic language for technical computing. arXiv:1209.5145v1.
[8]
Brisson, N., Launay, M., Mary, B., and Beaudoin, N. Conceptual basis, formalisations and parameterization of the STICS crop model. Quae éditions, 2009.
[9]
Campillo, F., and Rossi, V. Convolution Particle Filter for Parameter Estimation in General State-Space Models. IEEE Transactions in Aerospace and Electronics. 45, 3 (2009), 1063--1072.
[10]
Chen, Y. Bayesian Inference in Plant Growth Models for Prediction and Uncertainty Assessment. PhD thesis, Ecole Centrale Paris, 2014.
[11]
Chen, Y., and Cournède, P.-H. Data assimilation to reduce uncertainty of crop model prediction with convolution particle filtering. Ecological Modelling 290 (2014), 165--177.
[12]
Cournede, P.-H., Chen, Y., Wu, Q., Baey, C., and Bayol, B. Development and evaluation of plant growth models: methodology and implementation in the pygmalion platform. Mathematical Modelling of Natural Phenomena 8, 04 (2013), 112--130.
[13]
de Reffye, P., Heuvelink, E., Barthélémy, D., and Cournède, P.-H. Plant growth models. In Ecological Models. Vol. 4 of Encyclopedia of Ecology (5 volumes), S. Jorgensen and B. Fath, Eds. Elsevier, Oxford, 2008, 2824--2837.
[14]
Doucet, A., De Freitas, N., and Gordon, N. Sequential Monte Carlo methods in practice. Springer, New York, 2001.
[15]
Fearnhead, P. MCMC for state-space models. In Handbook of Markov chain Monte Carlo, S. Brooks et al., Eds. Chapman and Hall, London, 2011, 513--529.
[16]
Gilks, W. R. Markov chain monte carlo. In Encyclopedia of Biostatistics. John Wiley and Sons, Ltd, 2005.
[17]
Haario, H., and Tamminen, J. An adaptive Metropolis algorithm. Bernoulli 7 (2001), 223--242.
Index Terms
- A comparison of a generic MCMC-based algorithm for Bayesian estimation in C++, R and Julia: application to plant growth modeling
Recommendations
Bayesian estimation of the variance of a jitter using MCMC
SSAP '96: Proceedings of the 8th IEEE Signal Processing Workshop on Statistical Signal and Array Processing (SSAP '96)The problem treated in this paper is the Bayesian estimation of the variance of the sampling jitter occurring when a process is irregularly observed. This problem is often met in practice, and has already being examined using higher order statistics. ...
MCMC-based local parametric sensitivity estimations
Bayesian inferences for complex models need to be made by approximation techniques, mainly by Markov chain Monte Carlo (MCMC) methods. For these models, sensitivity analysis is a difficult task. A novel computationally low-cost approach to estimate ...
Sample selection for MCMC-based recommender systems
RecSys '13: Proceedings of the 7th ACM conference on Recommender systemsBayesian Inference with Markov Chain Monte Carlo (MCMC) has been shown to provide high prediction quality in recommender systems. The advantage over learning methods such as coordinate descent/alternating least-squares (ALS) or (stochastic) gradient ...
Comments
Information & Contributors
Information
Published In
Sponsors
- SIGSIM: ACM Special Interest Group on Simulation and Modeling
- SCS: Society for Modeling and Simulation International
Publisher
Society for Computer Simulation International
San Diego, CA, United States
Publication History
Published: 12 April 2015
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
SpringSim '15
Sponsor:
- SIGSIM
- SCS
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 55Total Downloads
- Downloads (Last 12 months)1
- Downloads (Last 6 weeks)0
Reflects downloads up to 18 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in