Posterior Distribution Of Parameters Research Articles

Copula Approximate Bayesian Computation Using Distribution Random Forests

Ongoing modern computational advancements continue to make it easier to collect increasingly large and complex datasets, which can often only be realistically analyzed using models defined by intractable likelihood functions. This Stats invited feature article introduces and provides an extensive simulation study of a new approximate Bayesian computation (ABC) framework for estimating the posterior distribution and the maximum likelihood estimate (MLE) of the parameters of models defined by intractable likelihoods, that unifies and extends previous ABC methods proposed separately. This framework, copulaABCdrf, aims to accurately estimate and describe the possibly skewed and high-dimensional posterior distribution by a novel multivariate copula-based meta-t distribution based on univariate marginal posterior distributions that can be accurately estimated by distribution random forests (drf), while performing automatic summary statistics (covariates) selection, based on robustly estimated copula dependence parameters. The copulaABCdrf framework also provides a novel multivariate mode estimator to perform MLE and posterior mode estimation and an optional step to perform model selection from a given set of models using posterior probabilities estimated by drf. The posterior distribution estimation accuracy of the ABC framework is illustrated and compared with previous standard ABC methods through several simulation studies involving low- and high-dimensional models with computable posterior distributions, which are either unimodal, skewed, or multimodal; and exponential random graph and mechanistic network models, each defined by an intractable likelihood from which it is costly to simulate large network datasets. This paper also proposes and studies a new solution to the simulation cost problem in ABC involving the posterior estimation of parameters from datasets simulated from the given model that are smaller compared to the potentially large size of the dataset being analyzed. This proposal is motivated by the fact that, for many models defined by intractable likelihoods, such as the network models when they are applied to analyze massive networks, the repeated simulation of large datasets (networks) for posterior-based parameter estimation can be too computationally costly and vastly slow down or prohibit the use of standard ABC methods. The copulaABCdrf framework and standard ABC methods are further illustrated through analyses of large real-life networks of sizes ranging between 28,000 and 65.6 million nodes (between 3 million and 1.8 billion edges), including a large multilayer network with weighted directed edges. The results of the simulation studies show that, in settings where the true posterior distribution is not highly multimodal, copulaABCdrf usually produced similar point estimates from the posterior distribution for low-dimensional parametric models as previous ABC methods, but the copula-based method can produce more accurate estimates from the posterior distribution for high-dimensional models, and, in both dimensionality cases, usually produced more accurate estimates of univariate marginal posterior distributions of parameters. Also, posterior estimation accuracy was usually improved when pre-selecting the important summary statistics using drf compared to ABC employing no pre-selection of the subset of important summaries. For all ABC methods studied, accurate estimation of a highly multimodal posterior distribution was challenging. In light of the results of all the simulation studies, this article concludes by discussing how the copulaABCdrf framework can be improved for future research.

Bayesian modeling and optimization for split-plot experiments with multiple responses

In many industrial processes, cost or time constraints make some input variables harder to change or control than others. An appropriate experimental design method is restricted randomization, which results in split-plot experiments. Empirical models that connect multiple quality characteristics with input variables play a crucial role in robust parameter design for split-plot experiments. At present, many modeling methods typically adopt the single response model for analyzing industrial processes in the split-plot experiments without considering correlation among multiple responses, correlation among whole plots, and uncertainty of model parameters. However, ignoring these issues can lead to poor product or process design. To solve these issues, this paper suggests a novel Bayesian modeling and optimization approach. We first construct a Bayesian multi-response linear mixed-effects model and obtain the posterior distribution for model parameters by employing Bayesian theorem. Then, the Gibbs sampling procedure is employed for the estimation of model parameters. Finally, the overall weighted desirability optimization function meeting the specification is developed to avoid acquiring ideal input settings with outliers. A simulation and an engineering case study demonstrate the validity of the proposed method. In comparison to existing methods, the optimization results given the proposed method are more robust and reliable.

Ensemble surrogate modeling of advective-dispersive transport with intraparticle diffusion model for column-leaching test

Column-leaching tests are a common approach for assessing the leaching behavior and resulting environmental risks of contaminated soils and waste materials, which are frequently reused for various construction purposes. The observed breakthrough curves of the contaminants are influenced by the complex dynamics of solute transport and kinetic inter-phase mass transfer. Disentangling these interactions necessitates numerical models. However, inverse modeling and sensitivity analysis can be time-consuming, especially when sorption kinetics are explicitly described by intraparticle diffusion, which requires discretizing the domain both in the flow direction along the column axis and inside the grains. To circumvent the need for such computationally intensive models, we have developed two different ensemble surrogate models. These models employ two separate ensemble methods: random forest stacking and inverse-distance weighted interpolation. Each method is applied to base surrogate models that cover different parts of the parameter space. The base surrogate models use the method of Extremely randomized Trees (ExtraTrees). The defined parameter range is based on the German standard for column-leaching tests. To optimize the base surrogate models, we utilized adaptive-sampling methods based on three distinct infill criteria: maximizing the expected improvement, staying within a certain Mahalanobis distance to the best estimate (both for exploitation), and maximizing the standard deviation (for exploration). The ensemble surrogate model demonstrates excellent performance in emulating the behavior of the original numerical model, with a relative root mean squared error of 0.09. We applied our proposed ensemble surrogate model to estimate the complete posterior parameter distribution using Simulation-Based Inference, specifically Neural Posterior Estimation, to determine the full parameter distribution conditioned on copper-leaching data from two different soils. Samples drawn from the posterior distribution align perfectly with the observed data for both the surrogate and original models.

Sequential and adaptive probabilistic integration for Bayesian model updating

The present study attempts to introduce a Bayesian perspective to marginal likelihood estimation. For this purpose, the Bayesian marginal likelihood inference (BMLI) framework is first developed, where the posterior statistics of the marginal likelihood are estimated using efficient random process sampling. The sequential and adaptive probabilistic integration (SAPI) method is then proposed within the BMLI framework to efficiently infer the posterior distribution of the model parameters as a result of the marginal likelihood estimation. In this method, an adaptive tempering of the likelihood function is presented to draw samples from a series of intermediate distributions to perform probabilistic integration sequentially, and the samples are finally converged to the target posterior domain. An active learning process is also designed to accelerate the sequential integration with a minimum number of likelihood function observations. Four numerical examples are investigated and the superiority of the SAPI method over other sequential sampling methods is demonstrated.

Integrating Bayesian inference and neural ODEs for microgrids dynamics parameters estimation

The integration of solar and wind energy sources in microgrids has witnessed significant growth, giving rise to distinct challenges due to their intermittent nature when it comes to achieving efficient microgrid control. However, estimating the parameters of the dynamic microgrid components facilitates capturing the complex and time-varying characteristics of renewable energy generation. This requires an accurate estimation of the parameters from the dynamic differential equations for effective modeling and control. In this research paper, we presented a novel methodology based on the integration of Bayesian inference and Neural ODEs. The Bayesian inference quantifies the uncertainty, and the Neural ODEs model the dynamic systems. By combining the strengths of both methods, we aimed to achieve a precise and robust parameter estimation of the dynamic microgrid components. The methodology is validated on a simulated microgrid that consists of a diesel generator, Solar PV array, double-fed induction generator, and a battery energy storage system. The results showed promised inferences estimation obtained from the parameter posterior distribution even in the presence of uncertainty. This can enhance our understanding of the dynamics of renewable energy systems and can contribute to the advancement of decision-making microgrid control strategies.

Practical guidelines for Bayesian phylogenetic inference using Markov chain Monte Carlo (MCMC)

Phylogenetic estimation is, and has always been, a complex endeavor. Estimating a phylogenetic tree involves evaluating many possible solutions and possible evolutionary histories that could explain a set of observed data, typically by using a model of evolution. Values for all model parameters need to be evaluated as well. Modern statistical methods involve not just the estimation of a tree, but also solutions to more complex models involving fossil record information and other data sources. Markov chain Monte Carlo (MCMC) is a leading method for approximating the posterior distribution of parameters in a mathematical model. It is deployed in all Bayesian phylogenetic tree estimation software. While many researchers use MCMC in phylogenetic analyses, interpreting results and diagnosing problems with MCMC remain vexing issues to many biologists. In this manuscript, we will offer an overview of how MCMC is used in Bayesian phylogenetic inference, with a particular emphasis on complex hierarchical models, such as the fossilized birth-death (FBD) model. We will discuss strategies to diagnose common MCMC problems and troubleshoot difficult analyses, in particular convergence issues. We will show how the study design, the choice of models and priors, but also technical features of the inference tools themselves can all be adjusted to obtain the best results. Finally, we will also discuss the unique challenges created by the incorporation of fossil information in phylogenetic inference, and present tips to address them.

Modeling Blazar Broadband Emission with Convolutional Neural Networks. II. External Compton Model

In the context of modeling spectral energy distributions (SEDs) for blazars, we extend the method that uses a convolutional neural network (CNN) to include external inverse Compton processes. The model assumes that relativistic electrons within the emitting region can interact with and up-scatter external photons originating from the accretion disk, the broad-line region, and the torus, to produce the observed high-energy emission. We trained the CNN on a numerical model that accounts for the injection of electrons, their self-consistent cooling, and pair creation-annihilation processes, considering both internal and all external photon fields. Despite the larger number of parameters compared to the synchrotron self-Compton model and the greater diversity in spectral shapes, the CNN enables an accurate computation of the SED for a specified set of parameters. The performance of the CNN is demonstrated by fitting the SED of two flat-spectrum radio quasars, namely 3C 454.3 and CTA 102, and obtaining their parameter posterior distributions. For the first source, the available data in the low-energy band allowed us to constrain the minimum Lorentz factor of the electrons, γmin , while for the second source, due to the lack of these data, γmin=102 was set. We used the obtained parameters to investigate the energetics of the system. The model developed here, along with one from Bégué et al., enables self-consistent, in-depth modeling of blazar broadband emissions within a leptonic scenario.

Simulation-based inference of developmental EEG maturation with the spectral graph model

The spectral content of macroscopic neural activity evolves throughout development, yet how this maturation relates to underlying brain network formation and dynamics remains unknown. Here, we assess the developmental maturation of electroencephalogram spectra via Bayesian model inversion of the spectral graph model, a parsimonious whole-brain model of spatiospectral neural activity derived from linearized neural field models coupled by the structural connectome. Simulation-based inference was used to estimate age-varying spectral graph model parameter posterior distributions from electroencephalogram spectra spanning the developmental period. This model-fitting approach accurately captures observed developmental electroencephalogram spectral maturation via a neurobiologically consistent progression of key neural parameters: long-range coupling, axonal conduction speed, and excitatory:inhibitory balance. These results suggest that the spectral maturation of macroscopic neural activity observed during typical development is supported by age-dependent functional adaptations in localized neural dynamics and their long-range coupling across the macroscopic structural network.

Bayesian peak identification method for low count rate gamma spectrum under short-time measurement based on physical priors

A Bayesian peak identification method based on physical model had been proposed in this study to perform qualitative analysis for the low count rate gamma spectrum under short-time measurement. Based on the prior knowledge about gamma detection, the distribution of energy deposited in the detector near the region of interest of a target peak through photoelectric effect, Compton scattering and multiple scattering had been analyzed and approximated using three elementary components, then, the distribution of each component after convolution had been studied, and at last, a probabilistic mixture model had been proposed to directly describe the distribution of measured energy within region of interest. When the data measurement has been completed, a Gibbs sampler was used to estimate the posterior distribution of model parameters to discriminate the existence of target peaks. The numerical and physical experiments were performed to verify the effectiveness of the parameter estimation algorithm and the performance of the proposed identification method. The numerical experiments showed the identification precision is proportional to counts and peak-to-total ratio, when the peak-to-total ratio is greater than 0.5, the proposed method could identify the peak with a probability greater than 50% given 50 observations. The physical experiment used a NaI(Tl) detector to measure 137Cs, 60Co and 152Eu sources in a laboratory environment at different equivalent distances, the peak identification precision had been tested for the proposed method, and the experiments proved that the probabilistic model is a good approximation to the physical process, and the proposed identification method outperformed the conventional method under short-time measurement conditions.

The role of Markov chain in Monte Carlo simulation

The Markov Chain Monte Carlo method has been applied in several fields and has thrived in academic research over the past few decades. The method is utilized to estimate the interest parameters posterior distribution through random sampling in the probability space. Monte Carlo method alone can accomplish random simulations for some complicated continuous integrals or discrete summations. However, the method requires the sample set corresponding to the distribution, which demands the application of Markov chain. Nevertheless, problems occur when the probability function of the samples remain unknown, under which circumstance the Markov Chain Monte Carlo method is no longer available. Therefore, the article introduced Metropolis-Hastings Algorithm as a resolution. The paper is intended to provide specific insights about Metropolis-Hastings Algorithm for Markov Chain Monte Carlo and the corresponding interdisciplinary applications. Specifically, the paper will discuss about using the theory mentioned above to check the relationships among species in a common ecosystem. The paper will cover several prepositive theories, including the Law of Large Numbers for Markov Chains, binary sequences, and Metropolis-Hastings Algorithm. Overall, specific cases and the way to draw correct conclusions will be most concerned.

Applying Machine Learning Techniques: Uncertainty Quantification in Nonlinear Dynamics Characters Predictions via Gated Recurrent Unit-Based Reduced-Order Models

The development of reduced-order models has been a pivotal advancement in the computational analysis of fluid dynamics, substantially simplifying the complexity and boosting the efficiency of simulations. The accuracy and practicality of these models largely depend on the reduction techniques applied and the inherent characteristics of the fluid dynamics systems they represent. In this paper, we introduce an innovative machine-learning framework for assessing model uncertainty in computationally intensive reduced-order models. By combining subspace construction methods with advanced Bayesian inference techniques, our approach effectively captures the posterior distribution of model parameters, thereby providing an accurate representation of uncertainty in aerodynamic performance predictions. We employ the NACA0012 airfoil as a case study to validate our method’s ability to enhance the efficiency of reduced-order models and precisely measure the uncertainty inherent in predictions made by recurrent neural networks. It is important to note that our approach is influenced by specific constraints and variables that significantly impact the mean and variability of the predicted final lift coefficient distribution. Our findings indicate that setting the goodness of fit (R2) threshold above 0.985 markedly improves the correlation between Computational Fluid Dynamics (CFD) outcomes and model predictions, increasing from 72.2% to 97.9% as the interval amplification factor adjusts from 1.5 to 3. However, this adjustment causes a considerable expansion of the confidence interval, from 0.0737 to 0.1282, an increase of over 70%. Despite these challenges, our machine learning-based methodology provides essential insights into the further development of reduced-order modeling and uncertainty quantification in fluid dynamics. This highlights the need for ongoing research into the model parameters, especially in applications concerning aircraft control systems, to meet design specifications and ensure system reliability.

Covariance-based MCMC for high-dimensional Bayesian updating with Sequential Monte Carlo

Sequential Monte Carlo (SMC) is a reliable method to generate samples from the posterior parameter distribution in a Bayesian updating context. The method samples a series of distributions sequentially, which start from the prior distribution and gradually approach the posterior distribution. Sampling from the distribution sequence is performed through application of a resample-move scheme, whereby the move step is performed using a Markov Chain Monte Carlo (MCMC) algorithm. The preconditioned Crank–Nicolson (pCN) is a popular choice for the MCMC step in high dimensional Bayesian updating problems, since its performance is invariant to the dimension of the prior distribution. This paper proposes two other SMC variants that use covariance information to inform the MCMC distribution proposals and compares their performance to the one of pCN-based SMC. Particularly, a variation of the pCN algorithm that employs covariance information, and the principle component Metropolis Hastings algorithm are considered. These algorithms are combined with an intermittent and recursive updating scheme of the target distribution covariance matrix based on the current MCMC progress. We test the performance of the algorithms in three numerical examples; a two dimensional algebraic example, the estimation of the flexibility of a cantilever beam and the estimation of the hydraulic conductivity field of an aquifer. The results show that covariance-based MCMC algorithms are capable of producing smaller errors in parameter mean and variance and better estimates of the model evidence compared to the pCN approach.

Toward Robust Corrections for Stellar Contamination in JWST Exoplanet Transmission Spectra

Transmission spectroscopy is still the preferred characterization technique for exoplanet atmospheres, although it presents unique challenges that translate into characterization bottlenecks when robust mitigation strategies are missing. Stellar contamination is one such challenge that can overpower the planetary signal by up to an order of magnitude, and thus not accounting for it can lead to significant biases in the derived atmospheric properties. Yet this accounting may not be straightforward, as important discrepancies exist between state-of-the-art stellar models and measured spectra and between models themselves. Here we explore the extent to which stellar models can be used to reliably correct for stellar contamination and yield a planet’s uncontaminated transmission spectrum. We find that discrepancies between stellar models can significantly contribute to the noise budget of JWST transmission spectra of planets around stars with heterogeneous photospheres, the true number of unique photospheric spectral components and their properties can only be accurately retrieved when the stellar models have sufficient fidelity, and under such optimistic circumstances the contribution of stellar contamination to the noise budget of a transmission spectrum is considerably below that of the photon noise for the standard transit observation setup. Therefore, we advocate for further development of model spectra of stars and their active regions in a data-driven manner, empirical approaches for deriving spectra of photospheric components using the observatories with which the atmospheric explorations are carried out, and analysis techniques accounting for multimodal posterior distributions for photospheric parameters of interest, which will be increasingly revealed by precise JWST measurements.

A Diffusion‐Based Uncertainty Quantification Method to Advance E3SM Land Model Calibration

AbstractCalibrating land surface models and accurately quantifying their uncertainty are crucial for improving the reliability of simulations of complex environmental processes. This, in turn, advances our predictive understanding of ecosystems and supports climate‐resilient decision‐making. Traditional calibration methods, however, face challenges of high computational costs and difficulties in accurately quantifying parameter uncertainties. To address these issues, we develop a diffusion‐based uncertainty quantification (DBUQ) method. Unlike conventional generative diffusion methods, which are computationally expensive and memory‐intensive, DBUQ innovates by formulating a parameterized generative model and approximates this model through supervised learning, which enables quick generation of parameter posterior samples to quantify its uncertainty. DBUQ is effective, efficient, and general‐purpose, making it suitable for site‐specific ecosystem model calibration and broadly applicable for parameter uncertainty quantification across various earth system models. In this study, we applied DBUQ to calibrate the Energy Exascale Earth System Model land model at the Missouri Ozark AmeriFlux forest site. Results indicated that DBUQ produced accurate parameter posterior distributions similar to those from Markov Chain Monte Carlo sampling but with 30 times less computing time. This significant improvement in efficiency suggests that DBUQ can enable rapid, site‐level model calibration at a global scale, enhancing our predictive understanding of climate impacts on terrestrial ecosystems.

Bayesian Model Updating of Multiscale Simulations Informing Corrosion Prognostics Using Conditional Invertible Neural Networks

Abstract Physics-based multi-scale corrosion simulation plays a vital role in predicting the evolution of pitting corrosion on large civil infrastructure such as miter gates, contributing to a model-informed structural health monitoring (SHM) strategy for risk-based asset health management. The physics-based analysis, however, may not accurately reflect the underlying true physics due to various uncertainty sources and needs to be updated using Bayesian inference methods based on observations to make the prediction closer to field observations. However, traditional Bayesian inference requires the evaluation of a likelihood function, which is often unavailable due to the complex model architecture and various surrogate models used in the analysis. Therefore, likelihood-free inference approaches are required for the updating of the multi-scale corrosion simulation models. This paper meets this need by proposing a conditional invertible neural network (cINN)-based Bayesian model updating method for an existing corrosion simulation model. We first train an cINN model based on simulated observations generated from a high-fidelity forward corrosion analysis model. A convolutional neural network (CNN)-based feature extraction algorithm is then employed to extract key features from corrosion images. After that, the extracted corrosion features from CNN is used as inputs of the cINN model to directly obtain posterior distributions of uncertain corrosion model parameters without evaluating the likelihood function. The results show that the proposed cINN-based model updating approach can provide more accurate inference results with a reduced computational cost in comparison to the classical approximate Bayesian computation (ABC) approach.

Unsupervised gas pipeline network leakage detection method based on improved graph deviation network

Historical data has shown that the natural gas pipeline network, as a critical urban lifeline system, is susceptible to disasters such as earthquakes resulting in leakage. Graph neural network modeling provides frontier rapid detection solutions for pipeline leakage; however, the challenges of collecting gas network anomaly data for training limit the precision and robustness of current model. This research proposes an unsupervised gas leakage detection and localization method based on a contained preprocessing process, with a graph deviation model that combines a structural learning approach with a graph neural network to model the spatial dependence of the sensors based on an attention mechanism, and variational inference models the posterior distributions of the hyper-parameters to optimize the model and improve the model precision. Meanwhile, in the preprocessing stage, the automatic optimization strategy of Northern Goshawk Optimization (NGO) for the best parameters K and ɑ in the Variable Mode Decomposition (VMD) efficiently extracts the valid signals and ensures that the model gives full play to the detection and localization effectiveness. The gas pipeline network dataset constructed by Pipeline Studio was used for comparing the performance of the model in this study with other frontier models. The results demonstrate that the model in this research has competitive detection Precision (93.31%), Recall (70.82%), and F1-score (0.81), and the posterior distribution of the model parameters strengthens the gas leakage localization precision, which provides a comprehensive solution for subsequent decision-making and reduces the hidden hazard of leakage effectively.

Practical guidelines for Bayesian phylogenetic inference using Markov Chain Monte Carlo (MCMC)

Phylogenetic estimation is, and has always been, a complex endeavor. Estimating a phylogenetic tree involves evaluating many possible solutions and possible evolutionary histories that could explain a set of observed data, typically by using a model of evolution. Modern statistical methods involve not just the estimation of a tree, but also solutions to more complex models involving fossil record information and other data sources. Markov Chain Monte Carlo (MCMC) is a leading method for approximating the posterior distribution of parameters in a mathematical model. It is deployed in all Bayesian phylogenetic tree estimation software. While many researchers use MCMC in phylogenetic analyses, interpreting results and diagnosing problems with MCMC remain vexing issues to many biologists. In this manuscript, we will offer an overview of how MCMC is used in Bayesian phylogenetic inference, with a particular emphasis on complex hierarchical models, such as the fossilized birth-death (FBD) model. We will discuss strategies to diagnose common MCMC problems and troubleshoot difficult analyses, in particular convergence issues. We will show how the study design, the choice of models and priors, but also technical features of the inference tools themselves can all be adjusted to obtain the best results. Finally, we will also discuss the unique challenges created by the incorporation of fossil information in phylogenetic inference, and present tips to address them.

Uncertainties quantification for damage localization in concrete based on Bayesian method

The presence of defects in concrete can diminish load-bearing capacity of structures, giving rise to potential concerns regarding safety and durability. Thus, a method that enhances the sensitivity, resolution and robustness of damage localization is critically necessary to assess the condition of concrete structures. This research presents a damage localization method based on Bayesian probabilistic fusion, and uncertainties from measurement and identification process are considered and quantified. The likelihood function is constructed based on the hyperbola-based damage localization method, and the posterior distributions of unknown parameters are calculated via Bayesian theorem combined with measurement data. Furthermore, a meso-level finite element model is established, wherein the concrete medium is considered as a three-phase composite material consisting of polygonal aggregates, mortar matrix and interface transition zones. Owing to the meso-level modeling, the propagation behavior of stress waves within concrete and complicated interactions between stress waves and concrete internal structures can be better characterized. Finally, the damage information, time-difference-of arrival, is extracted from the response signals and the efficiency of the proposed method is verified numerically. The numerical results demonstrate that the proposed probabilistic fusion method outperforms the conventional hyperbola-based method in terms of achieving high spatial resolution and resilience in damage localization.

A sparse factor model for clustering high-dimensional longitudinal data.

Recent advances in engineering technologies have enabled the collection of a large number of longitudinal features. This wealth of information presents unique opportunities for researchers to investigate the complex nature of diseases and uncover underlying disease mechanisms. However, analyzing such kind of data can be difficult due to its high dimensionality, heterogeneity and computational challenges. In this article, we propose a Bayesian nonparametric mixture model for clustering high-dimensional mixed-type (eg, continuous, discrete and categorical) longitudinal features. We employ a sparse factor model on the joint distribution of random effects and the key idea is to induce clustering at the latent factor level instead of the original data to escape the curse of dimensionality. The number of clusters is estimated through a Dirichlet process prior. An efficient Gibbs sampler is developed to estimate the posterior distribution of the model parameters. Analysis of real and simulated data is presented and discussed. Our study demonstrates that the proposed model serves as a useful analytical tool for clustering high-dimensional longitudinal data.

Improving crop yield estimation by unified model parameters and state variable with Bayesian inference

Data assimilation techniques integrating crop growth models and remote sensing technologies offer a feasible approach for large-scale crop yield estimation. Previous research has primarily focused on either recalibrate the uncertain model parameters or updating model state variables independently using remotely sensed observations. In this study, we developed a two-step inference algorithm that couples the parameter inference and the state update, to solve the joint posterior distribution of uncertain parameters and state variables given the remote sensing observations. An Observing Simulation System (OSS) experiment was first performed based on the WOFOST crop model to validate the effectiveness of the parameter inference method. The results indicate that, the parameter inference method successfully improved the estimation of different types of model parameters and enhanced yield estimation. Furthermore, leaf area index (LAI) retrieved from Sentinel-2 was assimilated into the WOFOST model to simulate winter wheat yield at the plot scale in the northeastern part of Henan Province. The results demonstrated that the proposed two-step inference algorithm can more effectively correct model simulation biases and improve winter wheat yield estimation accuracy (R²=0.58, MAPE=12.75 %, and RMSE=1112 kg·ha⁻¹), outperforming the standard EnKF algorithm (R²=0.51, MAPE=14.62 %, RMSE=1328 kg·ha⁻¹). Overall, attributed to its unified approach to estimating both model parameters and state variables, the proposed two-step inference algorithm shows promising application prospects for data assimilation-based crop yield estimation.