Bayesian Testing Research Articles

Modeling and Analysis of Language Transfer in Second Language Acquisition Assisted by Data Stream Fusion Algorithm

The Wiener process method is used to construct the degradation model, and the offline parameters are iteratively optimized using the EM algorithm combined with the training engine data. Based on Bayesian method and test engine data, an evolutionary data stream clustering algorithm (I-APDenStream) based on the fusion of nearest neighbor propagation and density is proposed. This algorithm uses the traditional two-stage processing model, namely online and offline clustering, to find the differences between English and Chinese general pronouns, the similarity and difference between the mother tongue and the target language, and the generalization of the sentence pattern or usage of the mother tongue. Use negative transfer. It aims to fully understand and scientifically grasp the status and role of language transfer in second language acquisition.

A new flexible Bayesian hypothesis test for multivariate data

A new flexible Bayesian hypothesis test for multivariate data

Bayesian estimation and testing in random-effects meta-analysis of rare binary events allowing for flexible group variability.

Rare binary events data arise frequently in medical research. Due to lack of statistical power in individual studies involving such data, meta-analysis has become an increasingly important tool for combining results from multiple independent studies. However, traditional meta-analysis methods often report severely biased estimates in such rare-event settings. Moreover, many rely on models assuming a pre-specified direction for variability between control and treatment groups for mathematical convenience, which may be violated in practice. Based on a flexible random-effects model that removes the assumption about the direction, we propose new Bayesian procedures for estimating and testing the overall treatment effect and inter-study heterogeneity. Our Markov chain Monte Carlo algorithm employs Pólya-Gamma augmentation so that all conditionals are known distributions, greatly facilitating computational efficiency. Our simulation shows that the proposed approach generally reports less biased and more stable estimates compared to existing methods. We further illustrate our approach using two real examples, one using rosiglitazone data from 56 studies and the other using stomach ulcers data from 41 studies.

Genetic analyses reveal a complex introduction history of the globally invasive tree Acacia longifolia

Acacia longifolia (Sydney golden wattle) is considered one of the most problematic plant invaders in Mediterranean-type ecosystems. In this study, we investigate the species’ invasion history by comparing the genetic diversity and structure of native (Australia) and several invasive range (Brazil, Portugal, South Africa, Spain, and Uruguay) populations and by modelling different introduction scenarios using these data. We sampled 272 A. longifolia individuals – 126 from different invasive ranges and 146 from the native range – from 41 populations. We genotyped all individuals at four chloroplast and 12 nuclear microsatellite markers. From these data we calculated diversity metrics, identified chloroplast haplotypes, and estimated population genetic structure based on Bayesian assignment tests. We used Approximate Bayesian Computation (ABC) models to infer the likely introduction history into each invaded country. In Australia, population genetic structure of A. longifolia appears to be strongly shaped by the Bass Strait and we identified two genetic clusters largely corresponding to mainland Australian and Tasmanian populations. We found invasive populations to represent a mixture of these clusters. Similar levels of genetic diversity were present in native and invasive ranges, indicating that invasive populations did not go through a genetic bottleneck. Bayesian assignment tests and chloroplast haplotype frequencies further suggested a secondary introduction event between South Africa and Portugal. However, ABC analyses could not confidently identify the native source(s) of invasive populations in these two countries, probably due to the known high propagule pressure that accompanied these introductions. ABC analyses identified Tasmania as the likely source of invasive populations in Brazil and Uruguay. A definitive native source for Spanish populations could also not be identified. This study shows that tracing the introduction history of A. longifolia is difficult, most likely because of the complexity associated with the extensive movement of the species around the world. Our findings should be considered when planning management and control efforts, such as biological control, in some invaded regions.

Mycobacterium tuberculosis complex detection in rural goat herds in South Africa using Bayesian latent class analysis

Animal tuberculosis affects a wide range of domestic and wild animal species, including goats (Capra hircus). In South Africa, Mycobacterium tuberculosis complex (MTBC) testing and surveillance in domestic goats is not widely applied, potentially leading to under recognition of goats as a potential source of M. bovis spread to cattle as well as humans and wildlife. The aim of this study was to estimate diagnostic test performance for four assays and determine whether M. bovis infection was present in goats sharing communal pastures with M. bovis positive cattle in the Umkhanyakude district of Northern Zululand, KwaZulu Natal. In 2019, 137 M. bovis-exposed goats were screened for MTBC infection with four diagnostic tests: the in vivo single intradermal comparative cervical tuberculin test (SICCT), in vitro QuantiFERON®-TB Gold (QFT) bovine interferon-gamma release assay (IGRA), QFT bovine interferon gamma induced protein 10 (IP-10) release assay (IPRA), and nasal swabs tested with the Cepheid GeneXpert® MTB/RIF Ultra (GXU) assay for detection of MTBC DNA. A Bayesian latent class analysis was used to estimate MTBC prevalence and diagnostic test sensitivity and specificity. Among the 137 M. bovis-exposed goats, positive test results were identified in 15/136 (11.0%) goats by the SICCT; 4/128 (3.1%) goats by the IPRA; 2/128 (1.6%) goats by the IGRA; and 26/134 (19.4%) nasal swabs by the GXU. True prevalence was estimated by our model to be 1.1%, suggesting that goats in these communal herds are infected with MTBC at a low level. Estimated posterior means across the four evaluated assays ranged from 62.7% to 80.9% for diagnostic sensitivity and from 82.9% to 97.9% for diagnostic specificity, albeit estimates of the former (diagnostic sensitivity) were dependent on model assumptions. The application of a Bayesian latent class analysis and multiple ante-mortem test results may improve detection of MTBC, especially when prevalence is low. Our results provide a foundation for further investigation to confirm infection in communal goat herds and identify previously unrecognized sources of intra- and inter-species transmission of MTBC.

Overgeneralizing emotions: Facial width-to-height revisited.

Facial width-to height ratio (fWHR), presumed to be shaped by testosterone during puberty, has been linked with aggressive, dominant, and power-seeking behavioral traits in adult males, although the causal mediation is still being disputed. To investigate the role of mere observer attribution bias in the association, we instructed participants to draw, feature-assemble, or photo-edit faces of fictitious males with aggressive-dominant character (compared with peaceloving-submissive), or powerful social status (compared with powerless). Across three studies involving 1,100 modeled faces in total, we observed little evidence for attribution bias with regards to facial width. Only in the photo-edited faces did character condition seem to affect fWHR; this difference, however, relied on displayed state emotions, not on static facial features. Anger, in particular, was expressed by lowered or V-shaped eyebrows, whereby facial height was reduced so that fWHR increased, relative to the comparison condition where the opposite happened. Using Bayesian analyses and equivalence testing, we confirmed that, in the absence of state emotionality, there was no effect of character condition on facial width. Our results add to a number of recent studies stressing the role of emotion overgeneralization in the association of fWHR with personality traits, an attributional bias that may give rise to a self-fulfilling prophecy. Methodologically, we infer that static images may be of limited use for investigations of fWHR because they cannot sufficiently differentiate between transient muscular activation and identity-related bone structures. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Deep Learning of GNSS Acquisition.

Signal acquisition is a crucial step in Global Navigation Satellite System (GNSS) receivers, which is typically solved by maximizing the so-called Cross-Ambiguity Function (CAF) as a hypothesis testing problem. This article proposes to use deep learning models to perform such acquisition, whereby the CAF is fed to a data-driven classifier that outputs binary class posteriors. The class posteriors are used to compute a Bayesian hypothesis test to statistically decide the presence or absence of a GNSS signal. The versatility and computational affordability of the proposed method are addressed by splitting the CAF into smaller overlapping sections, which are fed to a bank of parallel classifiers whose probabilistic results are optimally fused to provide a so-called probability ratio map from which acquisition is decided. Additionally, the article shows how noncoherent integration schemes are enabled through optimal data fusion, with the goal of increasing the resulting classifier accuracy. The article provides simulation results showing that the proposed data-driven method outperforms current CAF maximization strategies, enabling enhanced acquisition at medium-to-high carrier-to-noise density ratios.

Mitochondrial related genome-wide Mendelian randomization identifies putatively causal genes for multiple cancer types.

Mitochondrial dysfunction is a hallmark of cancer. However, it is unclear whether it is a cause of cancer. This two-sample Mendelian randomization (MR) analyses, uses genetic instruments to proxy the exposure of mitochondrial dysfunction and cancer summary statistics as outcomes, allowing for causal inferences. Summary statistics from 18 common cancers (2107-491,974 participants), gene expression, DNA methylation and protein expression quantitative trait loci (eQTL, mQTL and pQTL, respectively, 1000-31,684 participants) on individuals of European ancestry, were included. Genetic variants located within or close to the 1136 mitochondrial-related genes (in cis) and robustly associated with the mitochondrial molecular alterations were used as instrumental variables, and their causal associations with cancers were examined using summary-data-based MR (SMR) analyses. An additional five MR methods were used as sensitivity analyses to confirm the casual associations. A Bayesian test for colocalization between mitochondrial molecular QTLs and cancer risk loci was performed to provide insights into the potential regulatory mechanisms of risk variants on cancers. We identified potential causal relationships between mitochondrial-related genes and breast, prostate, gastric, lung cancer and melanoma by primary SMR analyses. The sensitivity and the colocalization analyses further refined four genes that have causal effects on three types of cancer. We found strong evidence of positive association of FDPS expression level with breast cancer risk (OR per SD, 0.66; 95% CI, 0.49-0.83; P=9.77×10-7), NSUN4 expression level with both breast cancer risk (OR per SD, 1.05; 95% CI, 1.03-1.07; P=5.24×10-6) and prostate cancer risk (OR per SD, 1.06; 95% CI, 1.03-1.09; P=1.01×10-5), NSUN4 methylation level with both breast and prostate cancer risk, and VARS2 methylation level with lung cancer risk. This data-driven MR study demonstrated the causal role of mitochondrial dysfunction in multiple cancers. Furthermore, this study identified candidate genes that can be the targets of potential pharmacological agents for cancer prevention. This work was supported by Styrelsen för Allmänna Sjukhusets i Malmö Stiftelse för bekämpande av cancer (20211025).

Examining contagion effects between global crude oil prices and the Southeast Asian stock markets during the COVID-19 pandemic

Many previous studies identify the contagion effect among various types of assets, defined as the increase in correlation of these assets during a financial or economic crisis. During the COVID-19 outbreak, a historic fall in global fuel demand and oil prices has been witnessed. Because crude oil has a strategic position among the export products of the Southeast Asian economies, even a tiny global oil price change leads to a plunge in these stock markets. This study addresses the spillovers of the volatility between the West Texas Intermediate crude oil prices and stock indices across six ASEAN emerging economies. Besides, the study examines whether a contagion connecting the global energy prices and these stock markets exists during the coronavirus pandemic. The empirical results are acquired by applying the Bayesian test for equality of means on the dynamic conditional correlations computed from DCC-GARCH models. The findings present positive volatility transmission from crude oil prices toward these emerging equity markets. During the health crisis, co-movements intensify, indicating the occurrence of contagion effects. The empirical results provide valid implications for policymakers and international investors because a precise volatility forecast is vital for managing portfolio risk. AcknowledgmentThis research is funded by University of Economics Ho Chi Minh City, Vietnam.

The complexity-aesthetics relationship for musical rhythm is more fixed than flexible: Evidence from children and expert dancers.

The urge to move to music (groove) depends in part on rhythmic syncopation in the music. For adults, the syncopation-groove relationship has an inverted-U shape: listeners want to move most to rhythms that have some, but not too much, syncopation. However, we do not know whether the syncopation-groove relationship is relatively sensitive to, or resistant to, a listener's experience. In two sets of experiments, we tested whether the syncopation-groove relationship is affected by dance experience or changes through development in childhood. Dancers and nondancers rated groove for 50 rhythmic patterns varying in syncopation. Dancers' and nondancers' ratings did not differ (and Bayesian tests provided substantial evidence that they were equivalent) in terms of mean groove and the optimal level of syncopation. Similarly, ballet and hip-hop dancers' syncopation-groove relationships did not differ. However, dancers had more robust syncopation-groove relationships (higher goodness-of-fit) than nondancers. Children (3-6 years old) completed two tasks to assess their syncopation-groove relationships: In a 2-alternative-forced choice task, children compared rhythms from 2 of 3 possible levels of syncopation (low, medium, and high) and chose which rhythm in a pair was better for dancing. In a dance task, children danced to the same rhythms. Results from both tasks indicated that for children, as for adults, medium syncopation rhythms elicit more groove than low syncopation rhythms. A follow-up experiment replicated the 2-alternative-forced choice task results. Taken together, the results suggest the optimal level of syncopation for groove is resistant to experience, although experience may affect the robustness of the inverted-U relationship. RESEARCH HIGHLIGHTS: In Experiment 1, dancers and nondancers rated groove (the urge to move) for musical rhythms, demonstrating the same inverted-U relationships between syncopation and groove. In Experiment 2, children and adults both chose rhythms with moderate syncopation more than low syncopation as more groove-inducing or better for dancing. Children also danced more for moderate than low syncopation, showing a close perception-behavior relationship across tasks. Similarities in the syncopation-groove relationship regardless of dance training and age suggest that this perceptual and behavioral groove response to rhythmic complexity may be quite resistant to experience.

Detection Framework of Abrupt Changes and Trends in Rainfall Erosivity in Three Gorges Reservoir, China

Rainfall erosivity is commonly used to estimate the probability of soil erosion caused by rainfall. The accurate detection of temporal changes in rainfall erosivity and the identification of abrupt changes and trends are important for understanding the physical causes of variation. In this study, a detection framework is introduced to identify temporal changes in rainfall erosivity time series as follows: (i) The significance of time series variation of rainfall erosivity is assessed based on the Hurst coefficient and divided into three levels: None, medium, and high. (ii) The detection of abrupt changes (Mann–Kendall, Moving T, and Bayesian tests) and trends (Spearman and Kendall rank correlation tests) of variate series and the correlation coefficient between the variation component and the original series is calculated. (iii) The modified series is obtained by preferentially eliminating the variation component (trend or change point) with larger correlation coefficients. (iv) We substituted the modified series into steps i to iii until the correlation coefficient was not significant. This framework is used to analyze the variation of rainfall erosivity in the Three Gorges Reservoir, China. The results showed that by using traditional methods, both an increasing trend and an upward change point were observed in Zigui station. However, after the upward change point was deducted from the annual rainfall erosivity series R(t), the resultant Rm(t) showed no statistically significant trend. Trend analysis should be performed considering the existence of an abrupt change to assess the long-term changes in rainfall erosivity series; otherwise, it would result in the wrong conclusion. In addition, the change points detected in the Rm(t) varied with the methods. Compared with the single-test method, the proposed framework can effectively reduce uncertainty.

Copula modelling with penalized complexity priors: the bivariate case

We explore the use of penalized complexity (PC) priors for assessing the dependence structure in a multivariate distribution F, with a particular emphasis on the bivariate case. We use the copula representation of F and derive the PC prior for the parameter governing the copula. We show that any alpha -divergence between a multivariate distribution and its counterpart with independent components does not depend on the marginal distribution of the components. This implies that the PC prior for the parameters of the copula can be elicited independently of the specific form of the marginal distributions. This represents a useful simplification in the model building step and may offer a new perspective in the field of objective Bayesian methodology. We also consider strategies for minimizing the role of subjective inputs in the prior elicitation step. Finally, we explore the use of PC priors in Bayesian hypothesis testing. Our prior is compared with competing default priors both for estimation purposes and testing.

An exploratory analysis of revision behavior development of L2 writers on an intensive English for academic purposes program using Bayesian methods

Revision is a fundamental part of the writing process and is particularly important in the production of high-quality academic writing. This study is an exploratory examination of changes in revision behavior, as measured by keystroke logging software, at the beginning (T1) and end (T2) of a one-month intensive English for Academic Purposes (EAP) course on n = 39 undergraduate and postgraduate students. Bayes Factors (BFs) are utilized as measures of strength of evidence for changes in behavior. In this paper, we examine the application of a Bayesian Hypothesis Testing (BHT) approach and its implications specifically for exploratory studies, i.e., studies with relatively small samples intended to search data for emergent patterns. The results show that, in most cases, we have moderate evidence against any change in behavior over time. Based on this evidence, we conclude that the experimental parameters of further exploratory work into the development of revisions should be modified to maximize the chance of finding patterns in the data from which to generate any confirmatory hypotheses.

Bayesian Log-Rank Test

Comparison of two survival curves is a fundamental problem in survival analysis. Although abundant frequentist methods have been developed for comparing survival functions, inference procedures from the Bayesian perspective are rather limited. In this article, we extract the quantity of interest from the classic log-rank test and propose its Bayesian counterpart. Monte Carlo methods, including a Gibbs sampler and a sequential importance sampling procedure, are developed to draw posterior samples of survival functions and a decision rule of hypothesis testing is constructed for making inference. Via simulations and real data analysis, the proposed Bayesian log-rank test is shown to be asymptotically equivalent to the classic one when noninformative prior distributions are used, which provides a Bayesian interpretation of the log-rank test. When using the correct prior information from historical data, the Bayesian log-rank test is shown to outperform the classic one in terms of power. R codes to implement the Bayesian log-rank test are also provided with step-by-step instructions.

Demand-Resource Profiles and Job Satisfaction in the Healthcare Sector: A Person-Centered Examination Using Bayesian Informative Hypothesis Testing.

Work characteristics may independently and jointly affect well-being, so that whether job demands deplete or energize employees depends on the resources available in the job. However, contradictory results on their joint effects have emerged so far in the literature. We argue that these inconsistencies can be partially explained by two arguments in the contemporary literature in the field. First, most studies in the job design domain are based on classic variable-centered methodologies which, although informative, are not well suited to investigate complex patterns of interactions among multiple variables. Second, these studies have mainly focused on generic work characteristics (e.g., workload, control, support), and are lacking in occupational specificity. Thus, to overcome these limitations, in the current research we include generic and occupation-specific work characteristics and adopt a person-centered approach to (a) identify different patterns of interactions of job demands and resources in a sample of healthcare employees, and (b) determine the degree to which these patterns are associated with employee well-being. We involved a sample of 1513 Italian healthcare providers and collected data on key job demands (workload, emotional dissonance, patient demands and physical demands) and resources (control, management support and peers' support). We focused on job satisfaction as a broad indicator of well-being. Latent profile analysis revealed four profiles of job demands and resources: high strain-isolated, resourceless, resourceful and active job on the ward. The results of Bayesian informative hypothesis testing showed the highest support for the hypothesis stating that healthcare employees belonging to the active job on the ward profile (medium-high demands, high resources) were the most satisfied. Conversely, employees belonging to the high strain-isolated profile (high demands, low resources) and the resourceless profile (medium-low demands, low resources) were the least satisfied. Overall, our study confirms the key role played by job resources in determining well-being in high-risk sectors, demonstrating that job satisfaction can develop both in challenging and less demanding situations. On a practical level, mapping the complexity of the healthcare psychosocial work environment has important implications, allowing for a better assessment process of employee well-being and helping to identify the most effective and fitting interventions.

Posterior contraction and testing for multivariate isotonic regression

We consider the nonparametric regression problem with multiple predictors and an additive error, where the regression function is assumed to be coordinatewise nondecreasing. We propose a Bayesian approach to make inferences on the multivariate monotone regression function, obtain the posterior contraction rate, and construct a universally consistent Bayesian testing procedure for multivariate monotonicity. To facilitate posterior analysis, we temporarily set aside the shape restrictions, and endow a prior on blockwise constant regression functions with independently normally distributed heights. The unknown variance of the error term is either estimated by the marginal maximum likelihood estimate, or is equipped with an inverse-gamma prior. Then the unrestricted block-heights are a posteriori also independently normally distributed given the error variance, by conjugacy. To comply with the shape restrictions, we project samples from the unrestricted posterior onto the class of multivariate monotone functions, inducing the “projection-posterior distribution”, to be used for making an inference. Under an L1-metric, we show that the projection-posterior based on n independent samples contracts around the true monotone regression function at the optimal rate n−1∕(2+d). Then we construct a Bayesian test for multivariate monotonicity based on the posterior probability of a shrinking neighborhood of the class of multivariate monotone functions. We show that the test is universally consistent, that is, the level of the Bayesian test goes to zero, and the power at any fixed alternative goes to one. Moreover, we show that for a smooth alternative function, power goes to one as long as its distance to the class of multivariate monotone functions is at least of the order of the estimation error for a smooth function. To the best of our knowledge, no other test for multivariate monotonicity is available in the Bayesian or the frequentist literature.

A general approximation to nested Bayes factors with informed priors

A staple of Bayesian model comparison and hypothesis testing Bayes factors are often used to quantify the relative predictive performance of two rival hypotheses. The computation of Bayes factors can be challenging, however, and this has contributed to the popularity of convenient approximations such as the Bayesian information criterion (BIC). Unfortunately, these approximations can fail in the case of informed prior distributions. Here, we address this problem by outlining an approximation to informed Bayes factors for a focal parameter . The approximation is computationally simple and requires only the maximum likelihood estimate and its standard error. The approximation uses an estimated likelihood of and assumes that the posterior distribution for is unaffected by the choice of prior distribution for the nuisance parameters. The resulting Bayes factor for the null hypothesis versus the alternative hypothesis is then easily obtained using the Savage–Dickey density ratio. Three real‐data examples highlight the speed and closeness of the approximation compared with bridge sampling and Laplace's method. The proposed approximation facilitates Bayesian reanalyses of standard frequentist results, encourages application of Bayesian tests with informed priors, and alleviates the computational challenges that often frustrate both Bayesian sensitivity analyses and Bayes factor design analyses. The approximation is shown to suffer under small sample sizes and when the posterior distribution of the focal parameter is substantially influenced by the prior distributions on the nuisance parameters. The proposed methodology may also be used to approximate the posterior distribution for under .

Improved Multitarget Tracking in the Presence of Port–Starboard Measurement Ambiguity Using the Bayes Factor

We consider an application where an unmanned underwater vehicle (UUV) equipped with an acoustic sensor seeks to estimate the location of surface ships using relative angle measurements to the ships. The estimation problem is challenging due to ships occasionally appearing and disappearing from the sensor's field of view. On occasion, poor geometry between the sensor and the ships, and port–starboard ambiguity that is inherent in the sensor contribute to the challenges in the estimation problem. The latter challenge arises because the sensor cannot distinguish between sound sources on its port and starboard side (port–starboard ambiguity). Therefore, every measurement is associated with two possible sound sources that map each relative angle to bearing projections on the port and starboard side of the UUV. We approach the problem of identifying the origin of sound sources (relative angle measurements) that are most likely to be of actual ships using Bayesian hypothesis testing. We propose an assignment method that uses the Bayes factor as the criteria to recursively associate measurements to the target that is most likely to be an actual ship and not a false target. The effectiveness of our approach is evaluated by comparing the performance of three multitarget tracking algorithms with and without our method. Performance is evaluated offline using sea-trial data captured by a UUV developed by the Naval Research Laboratory and Bluefin Robotics.

Spectrum Sensing for Combined GEO and LEO Satellites Based on Bi-LSTM and Bayesian Likelihood Ratio Test

目前的频谱感知算法以模型驱动为主，其感知性能过于依赖预定的统计模型，这使得其在信道环境复杂的卫星通信场景中的部署变得困难。对LEO卫星过境期间的信噪比波动情况进行分析，结果显示信噪比的波动达到14 dB。针对该复杂场景提出了一种基于双向长短期记忆网络及贝叶斯似然比检验联合的频谱感知算法。该算法不需要任何主信号的先验知识，可自动从主信号中学习隐藏特征并做出决策。基于Neyman-Pearson准则，在神经网络输出端设计了一种基于阈值的检测方案，可方便地控制恒定的虚警概率。仿真结果表明，所提算法在信噪比为–14 dB的情况下，仍能达到83%的检测性能，且始终优于卷积神经网络、多层感知机和基于模型驱动的能量检测算法。

Asset pricing models in South Africa: A comparative of regression analysis and the Bayesian approach

<abstract> <p>This study investigated the risk-return relationship using the Capital Asset Pricing Model (CAPM), Carhart four-factor Model (C4FM) and Fama and French Multifactor Models (FFMMs): F3FM and F5FM. This study analyzed the JSE ALSI returns of the South African market, in relation to the risk factors constructed by the data of twenty-six emerging markets, over the sample period from October 2000 to October 2021. The methodology employed was a comparison of the conventional regression analysis and novel Bayesian approach. Regression analysis estimated by Ordinary Least Squares (OLS) is one of the foremost methods used in South Africa. However, such an approach does not take into account the properties of the price data, namely, the asymmetric, volatile and random nature. While the Newey-West adjustment is one way to assist in capturing autocorrelation and volatility, it fails to consider asymmetry. The Bayesian approach accounts for all the aforementioned properties, overcoming the fundamental flaws of regression analysis. The additional model diagnostics highlighted in this study improved the Bayesian approach' robustness. The risk factors of the FFMMs estimated by regression analysis, with and without the Newey-West adjustment, were insignificant, whereas the Bayesian test results were significant. This finding clearly highlighted that model choice impacts the significance of parameter estimation and the financial decisions of investors, firms and policymakers. Jensen's alpha revealed CAPM to be optimal but none of the asset pricing models fully captured the risk premia. Thus, returns should be investigated with different risk measures for future research purposes.</p> </abstract>