Article
Physics, Multidisciplinary
Shaofan Liu, Shiliang Sun
Summary: Recently, the use of neural network parameterized flow models has been applied to design efficient Markov chain Monte Carlo (MCMC) transition kernels. However, the inefficient utilization of gradient information or the use of volume-preserving flows restricts their performance in sampling from multi-modal target distributions. In this paper, a novel training scheme is proposed, which divides the training process of transition kernels into exploration and training stages, allowing for full use of gradient information and the expressive power of deep neural networks. The proposed method achieves significant improvement in effective sample size and mixes quickly to the target distribution, outperforming other state-of-the-art parameterized transition kernels in various challenging distributions and real-world datasets.
Article
Economics
Nianling Wang, Zhusheng Lou
Summary: The stochastic volatility (SV) model is widely used to study time-varying volatility. However, the linearity assumption for transition equation in basic SV model is restrictive. To allow for nonlinearity, we proposed a semiparametric SV model that specifies a nonparametric transition equation for log-volatility using natural cubic splines. The empirical applications to Bitcoin and convertible bond return data indicate that the transition equations of their log-volatility are highly nonlinear. Taking nonlinearity into account, the semi-parametric SV model can improve the likelihood of the basic SV model both in-sample and out-of-sample.
ECONOMIC MODELLING
(2023)
Article
Computer Science, Artificial Intelligence
Sengul Dogan, Turker Tuncer
Summary: A new method using statistical decimal pattern and tunable Q-factor wavelet transform has been proposed in this study, which achieved high accuracy rates in surface electromyogram signal classification and outperformed other state-of-the-art methods according to the experimental results.
Review
Statistics & Probability
Christopher Nemeth, Paul Fearnhead
Summary: MCMC algorithms are considered the gold standard technique for Bayesian inference, but the computational cost can be prohibitive for large datasets, leading to the development of scalable Monte Carlo algorithms. One type of these algorithms is SGMCMC, which reduces per-iteration cost by utilizing data subsampling techniques.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2021)
Article
Engineering, Electrical & Electronic
Chenhao Li, Simon Godsill
Summary: The non-homogeneous Poisson process allows the intensity of point generation to vary across time or space domains, with applications in signal processing and machine learning but limited by intractable likelihood function and computationally efficient inference schemes. This paper proposes a framework that combines non-homogeneous Poisson model with continuous-time state-space models for efficient online inference. The proposed approach shows improved performance and computational efficiency compared to batch-based competitor algorithm and a simple baseline kernel estimation scheme.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2021)
Article
Medical Informatics
Kianoush Fathi Vajargah, Sara Ghaniyari Benis, Hamid Mottaghi Golshan
Summary: A new approach based on Monte Carlo Markov Chain sampling is proposed for noise removal from vital signals. Experimental results show that using Gaussian distribution sampling and retrieving the signal based on weighted average in selected samples allows for a more accurate estimation of the ideal signal, leading to an increase in the signal-to-noise ratio.
HEALTH INFORMATION SCIENCE AND SYSTEMS
(2021)
Article
Physics, Multidisciplinary
Zhikun Zhang, Min Dai, Xiangjun Wang
Summary: This study utilizes a mixed compound Poisson process by Markov switching model (CPMSM) to describe the non-stationarity of mixed stochastic jump processes. The results show that CPMSM is an effective method for describing jump behavior by controlling random counts through Markov chain switching.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2023)
Editorial Material
Economics
Mark Bognanni
Summary: This article introduces a method for fully Bayesian inference in the VAR-SV model and compares the different effects of using the triangular algorithm and the systemwide algorithm in the MCMC algorithm.
JOURNAL OF ECONOMETRICS
(2022)
Article
Engineering, Multidisciplinary
Ke Zhang, Kailun Su, Yunhan Yao, Qingsong Li, Suan Chen
Summary: This paper presents a dynamic evaluation model of Markov chain Monte Carlo (MCMC) roundness error measurement uncertainty based on a stochastic process. The model samples the stochastic process using the MCMC method and calculates the state transition function to reflect the autocorrelation characteristics of the parameters. A comparison between high-precision and low-precision measurements verifies the accuracy and stability of the model, showing that the MCMC method is consistent with the traditional GUM method and Monte Carlo method. The MCMC method based on the stochastic process achieves dynamic evaluation of roundness error measurement uncertainty, obtaining accurate results and improving the evaluation accuracy.
Review
Automation & Control Systems
Venkat Anantharam
Summary: This paper studies discrete time reversible Markov decision processes with finite state and action spaces, and the simplification of the policy iteration algorithm in such problems, as well as the relation between the finite time evolution of reward accumulation and the Gaussian free field associated to the controlled Markov chain.
SYSTEMS & CONTROL LETTERS
(2022)
Article
Engineering, Mechanical
P. L. Green, L. J. Devlin, R. E. Moore, R. J. Jackson, J. Li, S. Maskell
Summary: This paper discusses the optimization of the 'L-kernel' in Sequential Monte Carlo samplers to improve performance, resulting in reduced variance of estimates and fewer resampling requirements.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Review
Green & Sustainable Science & Technology
D. Hou, I. G. Hassan, L. Wang
Summary: Building Energy Model (BEM) calibration is crucial for accuracy, with recent focus on stochastic Bayesian inference calibration. However, confusion remains regarding theory, strengths, limitations, and implementations. Selecting appropriate mathematical models and tools poses a challenge.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2021)
Article
Physics, Multidisciplinary
Anna Pajor
Summary: This paper introduces a new method for estimating the Bayes factor, with simulation examples confirming its good performance. Additionally, it is found that the validity of reducing the hybrid MSV-MGARCH model to the MGARCH specification depends on the analyzed dataset and prior assumptions about model parameters.
Article
Computer Science, Interdisciplinary Applications
DanHua ShangGuan
Summary: The Monte Carlo method is a powerful tool in many research fields, but the increasing complexity of physical models and mathematical models requires efficient algorithms to overcome the computational cost.
JOURNAL OF COMPUTATIONAL PHYSICS
(2021)
Article
Pharmacology & Pharmacy
Leif-Thore Deck, David R. Ochsenbein, Marco Mazzotti
Summary: Freezing and freeze-drying processes are commonly used in pharmaceutical formulations to improve stability. However, batch heterogeneity can cause process failure. In this study, a modeling framework for large-scale freezing processes was developed and an open-source implementation was published. The model couples heat transfer with ice nucleation kinetics and showed how ice nucleation leads to heterogeneity. Various cooling protocols were investigated, and holding schemes were found to have similar solidification times as controlled nucleation, suggesting a potential pathway for process optimization.
INTERNATIONAL JOURNAL OF PHARMACEUTICS
(2022)