Bayesian Methodology Research Articles

Enhanced Bayesian RFI mitigation and transient flagging using likelihood reweighting

Abstract Contamination by radio frequency interference (RFI) is a ubiquitous challenge for radio astronomy. In particular, transient RFI is difficult to detect and avoid, especially in large data sets with many time bins. In this work, we present a Bayesian methodology for time-dependent, transient anomaly mitigation performed jointly with model fitting. The computation time for correcting transient anomalies in this manner in time-separated data sets grows proportionally with the number of time bins. We demonstrate that utilizing likelihood reweighting can allow our Bayesian anomaly mitigation method to be performed with a computation time close to independent of the number of time bins. In particular, we identify a factor of 44 improvement in computation time for a test case with 2000 time bins. We also demonstrate how this method enables the flagging threshold to be fit as a free parameter, fully automating the mitigation process. We find that this threshold fitting also prevents overcorrecting of the data in the case of wide priors. Finally, we investigate the potential of the methodology as a transient detector. We demonstrate that the method is able to reliably flag an individual anomalous data point out of 302 000 provided the Signal to Noise Ratio is $\ge\!{10}$.

Genomic selection for target traits in the Australian lentil breeding program.

Genomic selection (GS) uses associations between markers and phenotypes to predict the breeding values of individuals. It can be applied early in the breeding cycle to reduce the cross-to-cross generation interval and thereby increase genetic gain per unit of time. The development of cost-effective, high-throughput genotyping platforms has revolutionized plant breeding programs by enabling the implementation of GS at the scale required to achieve impact. As a result, GS is becoming routine in plant breeding, even in minor crops such as pulses. Here we examined 2,081 breeding lines from Agriculture Victoria's national lentil breeding program for a range of target traits including grain yield, ascochyta blight resistance, botrytis grey mould resistance, salinity and boron stress tolerance, 100-grain weight, seed size index and protein content. A broad range of narrow-sense heritabilities was observed across these traits (0.24-0.66). Genomic prediction models were developed based on 64,781 genome-wide SNPs using Bayesian methodology and genomic estimated breeding values (GEBVs) were calculated. Forward cross-validation was applied to examine the prediction accuracy of GS for these targeted traits. The accuracy of GEBVs was consistently higher (0.34-0.83) than BLUP estimated breeding values (EBVs) (0.22-0.54), indicating a higher expected rate of genetic gain with GS. GS-led parental selection using early generation breeding materials also resulted in higher genetic gain compared to BLUP-based selection performed using later generation breeding lines. Our results show that implementing GS in lentil breeding will fast track the development of high-yielding cultivars with increased resistance to biotic and abiotic stresses, as well as improved seed quality traits.

Hepatitis-related adverse events associated with immune checkpoint inhibitors in cancer patients: an observational, retrospective, pharmacovigilance study using the FAERS database.

Background: Immune checkpoint inhibitors (ICIs), including anti-PD-1, anti-PD-L1 and anti-CTLA-4 antibodies, have become a standard treatment for multiple cancer types. However, ICIs can induce immune-related adverse events, with hepatitis-related adverse events (HRAEs) being of particular concern. Our objective is to identify and characterize HRAEs that exhibit a significant association with ICIs using real-world data. Methods: In this observational and retrospective pharmacovigilance study, we extracted real-world adverse events reports from the FDA Adverse Event Reporting System database spanning from the first quarter of 2004 to the first quarter of 2023. We conducted both Frequentist and Bayesian methodologies in the framework of disproportionality analysis, which included the reporting odds ratios (ROR) and information components (IC) to explore the intricate relationship between ICIs and HRAEs. Results: Through disproportionality analysis, we identified three categories of HRAEs as being significantly related with ICIs, including autoimmune hepatitis (634 cases, ROR 19.34 [95% CI 17.80-21.02]; IC025 2.43), immune-mediated hepatitis (546 cases, ROR 217.24 [189.95-248.45]; IC025 4.75), and hepatitis fulminant (80 cases, ROR 4.56 [3.65-5.70]; IC025 0.49). The median age of patients who report ICI-related HRAEs was 63 years (interquartile range [IQR] 53.8-72), with a fatal outcome observed in 24.9% (313/1,260) of these reports. Cases pertaining to skin cancer, lung cancer, and kidney cancer constituted the majority of these occurrences. Patients treated with anti-PD-1 or anti-PD-L1 antibodies exhibited a higher frequency of immune-mediated hepatitis in comparison to those undergoing anti-CTLA-4 monotherapy, with a ROR of 3.59 (95% CI 1.78-6.18). Moreover, the dual ICI therapy demonstrated higher reporting rates of ICI-related HRAEs compared to ICI monotherapy. Conclusion: Our findings confirm that ICI treatment carries a significant risk of severe HRAEs, in particular autoimmune hepatitis, immune-mediated hepatitis, and hepatitis fulminant. Healthcare providers should exercise heightened vigilance regarding these risks when managing patients receiving ICIs.

E-Bayesian Estimation Using Spacing Function for Inverse Lindley Adaptive Type-I Progressively Censored Samples: Comparative Study with Applications.

For the first time, this paper offers the Bayesian and E-Bayesian estimation methods using the spacing function (SF) instead of the classical likelihood function. The inverse Lindley distribution, including its parameter and reliability measures, is discussed in this study through the mentioned methods, along with some other classical approaches. Six-point and six-interval estimations based on an adaptive Type-I progressively censored sample are considered. The likelihood and product of spacing methods are used in classical inferential setups. The approximate confidence intervals are discussed using both classical approaches. For various parameters, the Bayesian methodology is studied by taking both likelihood and SFs as observed data sources to derive the posterior distributions. Moreover, the E-Bayesian estimation method is considered by using the same data sources in the usual Bayesian approach. The Bayes and E-Bayes credible intervals using both likelihood and SFs are also taken into consideration. Several Monte Carlo experiments are carried out to assess the performance of the acquired estimators, depending on different accuracy criteria and experimental scenarios. Finally, two data sets from the engineering and physics sectors are analyzed to demonstrate the superiority and practicality of the suggested approaches.

Bayesian Methodologies with pyhf

bayesian_pyhf is a Python package that allows for the parallel Bayesian and frequentist evaluation of multi-channel binned statistical models. The Python library pyhf is used to build such models according to the HistFactory framework and already includes many frequentist inference methodologies. The pyhf-built models are then used as data-generating model for Bayesian inference and evaluated with the Python library PyMC. Based on Monte Carlo Chain Methods, PyMC allows for Bayesian modelling and together with the arviz library offers a wide range of Bayesian analysis tools.

Lead and delinquency rates; A spatio-temporal perspective.

Juvenile delinquency has significant social costs for perpetrators, victims, and communities. To understand the distribution of delinquency offenses this study considers the spatial clustering of juvenile delinquency with lead, race, and neighborhood deprivation using a longitudinal ecological design (N=4390) and a hierarchical model implemented in a Bayesian methodology that allows space-time interaction. The results show lead exposure is positively related to delinquency offense rates, and over time delinquency rates have become more concentrated in areas with higher levels of lead exposure and shares of Black or African American residents. The study emphasizes the isolation of neighborhoods with social problems and the importance of monitoring patterns of lead and crime at local levels as communities implement lead exposure mitigation programs.

Statistical inference of the Birnbaum-Saunders model using adaptive progressively hybrid censored data and its applications

<abstract><p>Lately, the Birnbaum-Saunders distribution has gained a lot of attention, mainly due to its different density shapes and the non-monotonicity property of its failure rates. This work considered some estimation issues for the Birnbaum-Saunders distribution using adaptive progressive Type-Ⅱ hybrid censoring. Point and interval estimations were performed employing both conventional and Bayesian methodologies. In addition to estimating the model parameters, we obtained point and interval estimates for the reliability and hazard rate functions. We looked at the method of maximum likelihood as a classical approach, and its asymptotic traits were employed to obtain approximate confidence ranges. From a Bayesian point of perspective, we considered the squared error loss function to obtain the point estimates of the various parameters. The Bayes and highest posterior density credible intervals were additionally determined. For the complex form of the posterior distribution, Bayes estimates and credible intervals were computed by sampling from the posterior distribution through the Markov chain Monte Carlo procedure. For assessing the performance of all of these estimators, a Monte Carlo simulation was employed. Several statistical standards were applied to check the effectiveness of various estimates for multiple levels of censoring with small, moderate, and large sample sizes. Finally, two scenarios for applications were given in order to highlight the usefulness of the supplied approaches.</p></abstract>

МЕТОДЫ, ТЕХНОЛОГИИ И СРЕДСТВА ИНТЕЛЛЕКТУАЛЬНОЙ ОБРАБОТКИ ИНФОРМАЦИИ О ДИНАМИКЕ МИГРАЦИИ В ТУЛЬСКОЙ ОБЛАСТИ

The purpose of the study is to assess and analyze the dynamics of the number of arrivals and departures from the Tula region to calculate the migration increase (decrease) of the population based on a fuzzy model built within the framework of the Bayesian intelligent measurement methodology.

Uncertain lithium-ion cathode kinetic decomposition modeling via Bayesian chemical reaction neural networks

Lithium-ion batteries are the focus of significant recent research interest due to their use in energy storage systems, electric vehicles, and other green technologies. Under various abuse conditions, these batteries undergo thermal runaway, which can lead to rapid temperature rise, flammable gas release, and fires. Current simulation approaches for thermal runaway at the full-cell scale involve utilizing kinetic models fit to experimental data from individual battery components. Despite substantial experimental uncertainty in the literature for these component experiments, prevalent models have not accounted for such uncertainty or noise. Here, we introduce uncertainty quantification for lithium-ion battery cathode thermal decomposition modeling via a novel Bayesian inference methodology. This approach leverages Chemical Reaction Neural Networks and particle-based uncertainty quantification to infer uncertain kinetic parameters while eliminating traditionally used simplifications, allowing for improvements to model accuracy and broader consideration of correlated parameters. We validated this new framework by learning an uncertain decomposition model for NCM333 (nickel-cobalt-manganese) cathode materials using differential scanning calorimetry (DSC) measurements with added synthetic noise. Then, we quantified the uncertainty in NCM811 cathode thermal runaway chemistry using experimental DSC measurements from various sources in the literature. Our methodology’s ability to account for correlated Arrhenius parameters led to much broader uncertain parameter ranges and thus more generalizability at higher temperatures. We additionally found that the NCM811 model distribution learned directly from experimental data in the literature has 4σ onset temperature ranges up to 20 °C wide and specific reaction enthalpy ranges accounting for upwards of 80% of the mean value, carrying significant implications for downstream applications. Our work bridges the gap between noisy or uncertain experimental data and practical cell-scale simulations, thereby facilitating more realistic and robust thermal runaway models that support enhanced battery safety and performance optimization.

Reliability analysis at usual operating settings for Weibull Constant-stress model with improved adaptive Type-Ⅱ progressively censored samples

<abstract><p>An improved adaptive Type-Ⅱ progressive censoring scheme was recently introduced to ensure that the examination duration will not surpass a specified threshold span. Employing this plan, this paper aimed to investigate statistical inference using Weibull constant-stress accelerated life tests. Two classical setups, namely maximum likelihood and maximum product of spacings, were explored to estimate the scale, shape, and reliability index under normal use conditions as well as their asymptotic confidence intervals. Through the same suggested classical setups, the Bayesian estimation methodology via the Markov chain Monte Carlo technique based on the squared error loss was considered to acquire the point and credible estimates. To compare the efficiency of the various offered approaches, a simulation study was carried out with varied sample sizes and censoring designs. The simulation findings show that the Bayesian approach via the likelihood function provides better estimates when compared with other methods. Finally, the utility of the proposed techniques was illustrated by analyzing two real data sets indicating the failure times of a white organic light-emitting diode and a pump motor.</p></abstract>

Bayes-xG: player and position correction on expected goals (xG) using Bayesian hierarchical approach.

This study employs Bayesian methodologies to explore the influence of player or positional factors in predicting the probability of a shot resulting in a goal, measured by the expected goals (xG) metric. Utilising publicly available data from StatsBomb, Bayesian hierarchical logistic regressions are constructed, analysing approximately 10,000 shots from the English Premier League (for the years of 2003 and 2015) to ascertain whether positional or player-level effects impact xG. The findings reveal positional effects in a basic model that includes only distance to goal and shot angle as predictors, highlighting that strikers and attacking midfielders exhibit a higher likelihood of scoring. However, these effects diminish when more informative predictors are introduced. Nevertheless, even with additional predictors, player-level effects persist, indicating that certain players possess notable positive or negative xG adjustments, influencing their likelihood of scoring a given chance. The study extends its analysis to data from Spain's La Liga ( K shots from 1973 and 2004 to 2020) and Germany's Bundesliga ( K shots from 2015), yielding comparable results. Additionally, the paper assesses the impact of prior distribution choices on outcomes, concluding that the priors employed in the models provide sound results but could be refined to enhance sampling efficiency for constructing more complex and extensive models feasibly.

Intensive tacrolimus trough level pharmacokinetic monitoring using Bayesian methodology preserves long-term kidney function after liver transplantation

Intensive tacrolimus trough level pharmacokinetic monitoring using Bayesian methodology preserves long-term kidney function after liver transplantation

Use of Bayesian Methods in the Process of Uranium Bioleaching by Acidithiobacillus ferrooxidans

This research is focused on investigating the utilization of Bayesian methodologies, specifically the Markov Chain Monte Carlo method, as well as filter sampling by importance and sequential resampling. The objective is to estimate kinetic parameters and state variables associated with the uranium bioleaching process by Acidithiobacillus ferrooxidans. Experimental data of cell concentration, uranium concentration, and concentrations of ferrous and ferric ions, obtained from literature, were employed. These measurements were evaluated using a mathematical model expressed by a system of ordinary differential equations. Three different mathematical models were evaluated, considering different uncertainties in experimental measurements and mathematical models (1% and 5%). The estimation results presented a good fit to the experimental data, with coefficients of determination in the range of 0.95 to 0.99. The validation of the mathematical models was obtained by reproducing the experimental measurements and the Bayesian techniques proved to be suitable for application in the bioleaching process.

Mapping the prevalence of cancer risk factors at the small area level in Australia

BackgroundCancer is a significant health issue globally and it is well known that cancer risk varies geographically. However in many countries there are no small area-level data on cancer risk factors with high resolution and complete reach, which hinders the development of targeted prevention strategies.MethodsUsing Australia as a case study, the 2017–2018 National Health Survey was used to generate prevalence estimates for 2221 small areas across Australia for eight cancer risk factor measures covering smoking, alcohol, physical activity, diet and weight. Utilising a recently developed Bayesian two-stage small area estimation methodology, the model incorporated survey-only covariates, spatial smoothing and hierarchical modelling techniques, along with a vast array of small area-level auxiliary data, including census, remoteness, and socioeconomic data. The models borrowed strength from previously published cancer risk estimates provided by the Social Health Atlases of Australia. Estimates were internally and externally validated.ResultsWe illustrated that in 2017–2018 health behaviours across Australia exhibited more spatial disparities than previously realised by improving the reach and resolution of formerly published cancer risk factors. The derived estimates revealed higher prevalence of unhealthy behaviours in more remote areas, and areas of lower socioeconomic status; a trend that aligned well with previous work.ConclusionsOur study addresses the gaps in small area level cancer risk factor estimates in Australia. The new estimates provide improved spatial resolution and reach and will enable more targeted cancer prevention strategies at the small area level. Furthermore, by including the results in the next release of the Australian Cancer Atlas, which currently provides small area level estimates of cancer incidence and relative survival, this work will help to provide a more comprehensive picture of cancer in Australia by supporting policy makers, researchers, and the general public in understanding the spatial distribution of cancer risk factors. The methodology applied in this work is generalisable to other small area estimation applications and has been shown to perform well when the survey data are sparse.

Bayes analysis of one-shot device testing data with correlated failure modes using copula models

Copula models are capable of modeling the dependence structure among the random variables, a phenomenon that is often required in the statistical analysis. Such models are the flexible substitutes of multivariate distributions because they model both the marginal distributions and the joint dependence structure distinctly. Because of such important features, the models are recognized as popular tools in a variety of situations including reliability engineering and survival analysis. The present paper studies a Bayesian approach using three Archimedean copulas, namely, the Gumbel Hougaard copula, the Frank copula and the Joe copula for analyzing one-shot device testing data with two correlated failure modes collected from a constant stress accelerated life test. One-shot devices are units that can be used only once and destroyed immediately after the use. Obviously, one obtains either left or right censored data on the failure times instead of actual failure times of the devices. Finally, all the considered copula models are compared using the Bayesian model selection tools. A real dataset is analyzed as an illustration of the proposed Bayesian methodology.

A practical guide to adopting Bayesian analyses in clinical research.

Bayesian statistical approaches are extensively used in new statistical methods but have not been adopted at the same rate in clinical and translational (C&T) research. The goal of this paper is to accelerate the transition of new methods into practice by improving the C&T researcher's ability to gain confidence in interpreting and implementing Bayesian analyses. We developed a Bayesian data analysis plan and implemented that plan for a two-arm clinical trial comparing the effectiveness of a new opioid in reducing time to discharge from the post-operative anesthesia unit and nerve block usage in surgery. Through this application, we offer a brief tutorial on Bayesian methods and exhibit how to apply four Bayesian statistical packages from STATA, SAS, and RStan to conduct linear and logistic regression analyses in clinical research. The analysis results in our application were robust to statistical package and consistent across a wide range of prior distributions. STATA was the most approachable package for linear regression but was more limited in the models that could be fitted and easily summarized. SAS and R offered more straightforward documentation and data management for the posteriors. They also offered direct programming of the likelihood making them more easily extendable to complex problems. Bayesian analysis is now accessible to a broad range of data analysts and should be considered in more C&T research analyses. This will allow C&T research teams the ability to adopt and interpret Bayesian methodology in more complex problems where Bayesian approaches are often needed.

Detecting Global Events with Bayesian Changepoint Detection on Flight Data

As a means of enhancing the intelligence community’s ability to be prepared and prompt in the detection of global events, we present a Bayesian changepoint detection methodology to detect anomalous activity in open-source flight data. We demonstrate a computationally inexpensive methodology to monitor a geographic location for changes in flight activity over time and demonstrate its utility in a case study of the 2022 Russian invasion of Ukraine. Beyond flight data, the methodology described in this paper shows the potential to generate near real-time situational awareness using open-source time series data streams. This paper was previously published and presented in the Donald R. Keith Memorial Capstone Conference at USMA in May of 2023.

MSR48 Using Bayesian Methodology to Minimize Control Recruitment in Trials of Relapsed and Refractory (RR) Multiple Myeloma (MM) Through Incorporation of Prior Control Response Information Borrowed From Real-World Evidence

MSR48 Using Bayesian Methodology to Minimize Control Recruitment in Trials of Relapsed and Refractory (RR) Multiple Myeloma (MM) Through Incorporation of Prior Control Response Information Borrowed From Real-World Evidence

Scalable Inverse Uncertainty Quantification by Hierarchical Bayesian Modeling and Variational Inference

Inverse Uncertainty Quantification (IUQ) has gained increasing attention in the field of nuclear engineering, especially nuclear thermal-hydraulics (TH), where it serves as an important tool for quantifying the uncertainties in the physical model parameters (PMPs) while making the model predictions consistent with the experimental data. In this paper, we present an extension to an existing Bayesian inference-based IUQ methodology by employing a hierarchical Bayesian model and variational inference (VI), and apply this novel framework to a real-world nuclear TH scenario. The proposed approach leverages a hierarchical model to encapsulate group-level behaviors inherent to the PMPs, thereby mitigating existing challenges posed by the high variability of PMPs under diverse experimental conditions and the potential overfitting issues due to unknown model discrepancies or outliers. To accommodate computational scalability and efficiency, we utilize VI to enable the framework to be used in applications with a large number of variables or datasets. The efficacy of the proposed method is evaluated against a previous study where a No-U-Turn-Sampler was used in a Bayesian hierarchical model. We illustrate the performance comparisons of the proposed framework through a synthetic data example and an applied case in nuclear TH. Our findings reveal that the presented approach not only delivers accurate and efficient IUQ without the need for manual tuning, but also offers a promising way for scaling to larger, more complex nuclear TH experimental datasets.

A dynamic power prior approach to non-inferiority trials for normal means.

Non-inferiority trials compare new experimental therapies to standard ones (active control). In these experiments, historical information on the control treatment is often available. This makes Bayesian methodology appealing since it allows a natural way to exploit information from past studies. In the present paper, we suggest the use of previous data for constructing the prior distribution of the control effect parameter. Specifically, we consider a dynamic power prior that possibly allows to discount the level of borrowing in the presence of heterogeneity between past and current control data. The discount parameter of the prior is based on the Hellinger distance between the posterior distributions of the control parameter based, respectively, on historical and current data. We develop the methodology for comparing normal means and we handle the unknown variance assumption using MCMC. We also provide a simulation study to analyze the proposed test in terms of frequentist size and power, as it is usually requested by regulatory agencies. Finally, we investigate comparisons with some existing methods and we illustrate an application to a real case study.