scholarly journals Bayesian multi-scale modeling for aggregated disease mapping data

2015 ◽  
Vol 26 (6) ◽  
pp. 2726-2742 ◽  
Author(s):  
Mehreteab Aregay ◽  
Andrew B Lawson ◽  
Christel Faes ◽  
Russell S Kirby
Keyword(s):  
Random Effect ◽  
Disease Mapping ◽  
Deviance Information Criterion ◽  
Simulated Data ◽  
Information Criterion ◽  
Modeling Framework ◽  
Multi Scale ◽  
Shared Random Effects ◽  
Convolution Model ◽  
Resolution Level

In disease mapping, a scale effect due to an aggregation of data from a finer resolution level to a coarser level is a common phenomenon. This article addresses this issue using a hierarchical Bayesian modeling framework. We propose four different multiscale models. The first two models use a shared random effect that the finer level inherits from the coarser level. The third model assumes two independent convolution models at the finer and coarser levels. The fourth model applies a convolution model at the finer level, but the relative risk at the coarser level is obtained by aggregating the estimates at the finer level. We compare the models using the deviance information criterion (DIC) and Watanabe-Akaike information criterion (WAIC) that are applied to real and simulated data. The results indicate that the models with shared random effects outperform the other models on a range of criteria.

scholarly journals Random scaling factors in Bayesian distributional regression models with an application to real estate data

2019 ◽  
Vol 20 (4) ◽  
pp. 347-368
Author(s):  
Alexander Razen ◽  
Stefan Lang
Keyword(s):  
Real Estate ◽  
Random Effect ◽  
Deviance Information Criterion ◽  
Information Criterion ◽  
Simultaneous Estimation ◽  
Response Distribution ◽  
Scaling Factors ◽  
Spatially Correlated ◽  
Non Linear ◽  
Different Response

Distributional structured additive regression provides a flexible framework for modelling each parameter of a potentially complex response distribution in dependence of covariates. Structured additive predictors allow for an additive decomposition of covariate effects with non-linear effects and time trends, unit- or cluster-specific heterogeneity, spatial heterogeneity and complex interactions between covariates of different type. Within this framework, we present a simultaneous estimation approach for multiplicative random effects that allow for cluster-specific heterogeneity with respect to the scaling of a covariate′s effect. More specifically, a possibly non-linear function f( z) of a covariate z may be scaled by a multiplicative and possibly spatially correlated cluster-specific random effect (1+αc). Inference is fully Bayesian and is based on highly efficient Markov Chain Monte Carlo (MCMC) algorithms. We investigate the statistical properties of our approach within extensive simulation experiments for different response distributions. Furthermore, we apply the methodology to German real estate data where we identify significant district-specific scaling factors. According to the deviance information criterion, the models incorporating these factors perform significantly better than standard models without (spatially correlated) random scaling factors.

scholarly journals PENERAPAN MODEL SPATIAL LOGIT-NORMAL PADA SMALL AREA ESTIMATION DENGAN METODE HIERARCHICAL BAYES

2020 ◽  
Vol 2019 (1) ◽  
pp. 59-66
Author(s):  
Taly Purwa
Keyword(s):  
Monte Carlo ◽  
Markov Chain ◽  
Small Area ◽  
Small Area Estimation ◽  
Random Effect ◽  
Deviance Information Criterion ◽  
Information Criterion ◽  
Hierarchical Bayes ◽  
Area Estimation ◽  
Spatial Random Effect

Penelitian ini menerapkan model Spatial Logit-normal pada Small Area Estimation (SAE) untuk estimasi proporsi penduduk dengan asupan kalori minimum di bawah 1.400 kkal/kapita/hari pada level kecamatan di Provinsi Bali Tahun 2014 yang merupakan indikator 2.1.2(A) pada tujuan ke-2 SDGs dalam rangka mengukur capaian dan mendukung tercapainya target SDGs pada level lebih tinggi. Terdapat tiga model SAE yang digunakan dengan spesifikasi random effect yang berbeda, yaitu model dengan random effect yang bersifat saling bebas (independen), spatial random effect (iCAR) serta model dengan kedua jenis random effect sekaligus (BYM). Penggunaan unsur spatial random effect diharapkan dapat meningkatkan efisiensi hasil estimasi. Metode estimasi menggunakan pendekatan Hierarchical Bayes (HB) dengan metode Markov Chain Monte Carlo (MCMC) algoritma Gibbs Sampling. Estimasi parameter pada ketiga model menunjukkan hasil yang relatif tidak berbeda dimana hanya ada satu variabel prediktor yang memiliki pengaruh signifikan, yaitu proporsi keluarga pertanian, pada model dengan random effect independen dan model BYM. Sedangkan pada model iCAR tidak ada satu pun variabel prediktor yang berpengaruh signifikan. Berdasarkan nilai Deviance Information Criterion (DIC), model terbaik adalah model BYM. Akan tetapi penambahan unsur spatial random effect bersamaan dengan random effect independen tidak secara signifikan dapat meningkatkan efisiensi hasil estimasi akibat dari minimnya nilai dependensi spasial Moran’s I. Secara visual, pemetaan hasil estimasi dengan model terbaik tidak menunjukkan adanya pola persebaran atau pengelompokan tertentu pada level kecamatan.

scholarly journals Use of Bayesian Markov Chain Monte Carlo Methods to Model Kuwait Medical Genetic Center Data: An Application to Down Syndrome and Mental Retardation

Mathematics ◽  
2021 ◽  
Vol 9 (3) ◽  
pp. 248
Author(s):  
Reem Aljarallah ◽  
Samer A Kharroubi
Keyword(s):  
Monte Carlo ◽  
Down Syndrome ◽  
Mental Retardation ◽  
Markov Chain ◽  
Markov Chain Monte Carlo ◽  
Univariate Analysis ◽  
Deviance Information Criterion ◽  
Predictive Ability ◽  
Information Criterion ◽  
Medical Genetic

Logit, probit and complementary log-log models are the most widely used models when binary dependent variables are available. Conventionally, these models have been frequentists. This paper aims to demonstrate how such models can be implemented relatively quickly and easily from a Bayesian framework using Gibbs sampling Markov chain Monte Carlo simulation methods in WinBUGS. We focus on the modeling and prediction of Down syndrome (DS) and Mental retardation (MR) data from an observational study at Kuwait Medical Genetic Center over a 30-year time period between 1979 and 2009. Modeling algorithms were used in two distinct ways; firstly, using three different methods at the disease level, including logistic, probit and cloglog models, and, secondly, using bivariate logistic regression to study the association between the two diseases in question. The models are compared in terms of their predictive ability via R2, adjusted R2, root mean square error (RMSE) and Bayesian Deviance Information Criterion (DIC). In the univariate analysis, the logistic model performed best, with R2 (0.1145), adjusted R2 (0.114), RMSE (0.3074) and DIC (7435.98) for DS, and R2 (0.0626), adjusted R2 (0.0621), RMSE (0.4676) and DIC (23120) for MR. In the bivariate case, results revealed that 7 and 8 out of the 10 selected covariates were significantly associated with DS and MR respectively, whilst none were associated with the interaction between the two outcomes. Bayesian methods are more flexible in handling complex non-standard models as well as they allow model fit and complexity to be assessed straightforwardly for non-nested hierarchical models.

scholarly journals Boosting heritability: estimating the genetic component of phenotypic variation with multiple sample splitting

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
The Tien Mai ◽  
Paul Turner ◽  
Jukka Corander
Keyword(s):  
Random Effect ◽  
Simulated Data ◽  
Fixed Effect Model ◽  
Accurate Estimate ◽  
Limited Attention ◽  
Effect Model ◽  
Generic Strategy ◽  
Multiple Sample ◽  
Inference Strategy ◽  
Resistance Data

Abstract Background Heritability is a central measure in genetics quantifying how much of the variability observed in a trait is attributable to genetic differences. Existing methods for estimating heritability are most often based on random-effect models, typically for computational reasons. The alternative of using a fixed-effect model has received much more limited attention in the literature. Results In this paper, we propose a generic strategy for heritability inference, termed as “boosting heritability”, by combining the advantageous features of different recent methods to produce an estimate of the heritability with a high-dimensional linear model. Boosting heritability uses in particular a multiple sample splitting strategy which leads in general to a stable and accurate estimate. We use both simulated data and real antibiotic resistance data from a major human pathogen, Sptreptococcus pneumoniae, to demonstrate the attractive features of our inference strategy. Conclusions Boosting is shown to offer a reliable and practically useful tool for inference about heritability.

A novel synergistic multi-scale modeling framework to predict micro- and meso-scale damage behaviors of 2D triaxially braided composite

2021 ◽  
pp. 105678952110339
Author(s):  
Hongyong Jiang ◽  
Yiru Ren ◽  
Qiduo Jin
Keyword(s):  
Compressive Load ◽  
Scale Model ◽  
Modeling Framework ◽  
Braided Composite ◽  
Multi Scale ◽  
Scale Modeling ◽  
Bridge Model ◽  
Multi Scale Modeling ◽  
Transverse Damage ◽  
Meso Scale

A novel synergistic multi-scale modeling framework with a coupling of micro- and meso-scale is proposed to predict damage behaviors of 2D-triaxially braided composite (2DTBC). Based on the Bridge model, the internal stress and micro damage of constituent materials are respectively coupled with the stress and damage of tow. The initial effective elastic properties of tow (IEEP) used as the predefined data are estimated by micro-mechanics models. Due to in-situ effects, stress concentration factor (SCF) is considered in the micro matrix, exhibiting progressive damage accumulation. Comparisons of IEEP and strengths between the Bridge and Chamis’ theory are conducted to validate the values of IEEP and SCF. Based on the representative volume element (RVE), the macro properties and damage modes of 2DTBC are predicted to be consistent with available experiments and meso-scale simulation. Both axial and transverse damage mechanisms of 2DTBC under tensile or compressive load are revealed. Micro fiber and matrix damage accumulations have significant effects on the meso-scale axial and transverse damage of tows due to multi-scale coupling effects. Different from existing meso-/multi-scale models, the proposed multi-scale model can capture a crucial phenomenon that the transverse damage of tow is vulnerable to micro fiber fracture. The proposed multi-scale framework provides a robust tool for future systematic studies on constituent materials level to larger-scale aeronautical materials.

PSXV-9 Transgenerational epigenetic variance for production and reproduction traits in maternal-line pigs

2021 ◽  
Vol 99 (Supplement_3) ◽  
pp. 258-259
Author(s):  
Jason R Graham ◽  
Jay S Johnson ◽  
Andre C Araujo ◽  
Jeremy T Howard ◽  
Luiz F Brito
Keyword(s):  
Additive Genetic Variance ◽  
Phenotypic Expression ◽  
Deviance Information Criterion ◽  
Information Criterion ◽  
Pedigree Information ◽  
Relationship Matrix ◽  
Maternal Line ◽  
Genetic Heritability ◽  
Reproduction Traits ◽  
Search Approach

Abstract Modeling epigenetic factors impacting phenotypic expression of economically important traits has become a hot-topic in the field of animal breeding due to the variability in genetic expression caused by environmental stressors (e.g., heat stress). This variability may be due, in part, to in-utero epigenomic remodeling, which has been reported to be passed from parent to offspring. We aimed to estimate transgenerational epigenetic variance for various production and reproduction traits measured in a maternal-line pig population, using a Bayesian approach. The phenotypes for production [n = 10,862; i.e., weaning weight (WW), birth weight (BW) and ultrasound-backfat thickness (BF)] and reproduction [n = 5,235, i.e., number of piglets born alive (NBA) and total number of piglets born (TB)] traits from a purebred Landrace population were provided by Smithfield Premium Genetics (NC, USA). The pedigree information traced back to 10 generations. Single-trait genetic analyses were performed using mixed models that included additive genetic, common environmental, and epigenetic random effects. The Gibbs sampler algorithm based on Markov chain Monte Carlo was used to estimate the variance components. The epigenetic relationship matrix was constructed using a recursive parameter (λ) related to the transmissibility coefficient of epigenetic markers. A grid search approach was used to define the optimal λ value (λ values ranged from 0.1 to 0.5, with an interval of 0.1). The optimal λ value was determined based on the deviance information criterion, and it was used to estimate the additive and epigenetic variances. For instance, based on preliminary results, the optimal λ value estimated for TB was 0.3 with an additive genetic variance of 0.94 (0.19 PSD) and epigenetic variance of 0.67 (0.18 PSD). The additive genetic heritability was 0.076 (0.015 PSD) and the estimated epigenetic heritability was 0.053 (0.015 PSD). This preliminary result suggests that epigenetics contribute to the non-Mendelian variability in pigs.

scholarly journals A multi-scale multi-physics modeling framework of laser powder bed fusion additive manufacturing process

2018 ◽  
Vol 73 (3) ◽  
pp. 151-157 ◽  
Author(s):  
Jing Zhang ◽  
Yi Zhang ◽  
Weng Hoh Lee ◽  
Linmin Wu ◽  
Hyun-Hee Choi ◽  
...  
Keyword(s):  
Additive Manufacturing ◽  
Manufacturing Process ◽  
Powder Bed Fusion ◽  
Laser Powder Bed Fusion ◽  
Modeling Framework ◽  
Powder Bed ◽  
Multi Scale ◽  
Physics Modeling

scholarly journals Utilização da modelagem inteiramente bayesiana na detecção de padrões de variação de risco relativo de mortalidade infantil no Rio Grande do Sul, Brasil

2009 ◽  
Vol 25 (7) ◽  
pp. 1501-1510 ◽  
Author(s):  
Sérgio Kakuta Kato ◽  
Diego de Matos Vieira ◽  
Jandyra Maria Guimarães Fachel
Keyword(s):  
Deviance Information Criterion ◽  
Rio Grande ◽  
Information Criterion ◽  
Rio Grande Do Sul ◽  
Standardised Mortality Ratio ◽  
Mortality Ratio

Neste artigo são analisados os fatores possivelmente associados à mortalidade infantil nos 496 municípios do Rio Grande do Sul, Brasil, com base em dados acumuladas entre os anos de 2001 a 2004, obtidos pela análise de regressão utilizando modelagem inteiramente bayesiana como alternativa para superar a autocorrelação espacial e a instabilidade dos estimadores clássicos, como a taxa bruta e a SMR (Standardised Mortality Ratio). Foram comparadas diferentes especificações de componente espacial e covariáveis, provenientes dos blocos do Índice de Desenvolvimento Sócio-econômico da Fundação de Economia e Estatística (IDESE/FEE-2003). Verificou-se que o modelo que utiliza a estrutura espacial além da covariável educação apresenta melhor desempenho, quando comparado pelo critério DIC (Deviance Information Criterion). Comparando as estimativas das SMR com os riscos relativos obtidos pela modelagem inteiramente bayesiana, foi possível observar um ganho substancial na interpretação e na detecção de padrões de variação do risco de mortalidade infantil nos municípios do Rio Grande do Sul ao utilizar essa modelagem. A região da Serra Gaúcha destacou-se com baixo risco relativo e estimativas muito homogêneas.

Understanding historical summer flounder (Paralichthys dentatus) abundance patterns through the incorporation of oceanography-dependent vital rates in Bayesian hierarchical models

2019 ◽  
Vol 76 (8) ◽  
pp. 1275-1294 ◽  
Author(s):  
Cecilia A. O’Leary ◽  
Timothy J. Miller ◽  
James T. Thorson ◽  
Janet A. Nye
Keyword(s):  
Population Dynamics ◽  
Gulf Stream ◽  
Stock Assessment ◽  
Deviance Information Criterion ◽  
Information Criterion ◽  
Natural Mortality ◽  
Vital Rates ◽  
Summer Flounder ◽  
Paralichthys Dentatus ◽  
Abundance Patterns

Climate can impact fish population dynamics through changes in productivity and shifts in distribution, and both responses have been observed for many fish species. However, few studies have incorporated climate into population dynamics or stock assessment models. This study aimed to uncover how past variations in population vital rates and fishing pressure account for observed abundance variation in summer flounder (Paralichthys dentatus). The influences of the Gulf Stream Index, an index of climate variability in the Northwest Atlantic, on abundance were explored through natural mortality and stock–recruitment relationships in age-structured hierarchical Bayesian models. Posterior predictive loss and deviance information criterion indicated that out of tested models, the best estimates of summer flounder abundances resulted from the climate-dependent natural mortality model that included log-quadratic responses to the Gulf Stream Index. This climate-linked population model demonstrates the role of climate responses in observed abundance patterns and emphasizes the complexities of environmental effects on populations beyond simple correlations. This approach highlights the importance of modeling the combined effect of fishing and climate simultaneously to understand population dynamics.

scholarly journals Boosting Heritability: Estimating the Genetic Component of Phenotypic Variation With Multiple Sample Splitting

2021 ◽  
Author(s):  
The Tien Mai ◽  
Paul Turner ◽  
Jukka Corander
Keyword(s):  
Random Effect ◽  
Simulated Data ◽  
Fixed Effect Model ◽  
Limited Attention ◽  
Reliable Estimate ◽  
Effect Model ◽  
Generic Strategy ◽  
Multiple Sample ◽  
Inference Strategy ◽  
Resistance Data

Abstract Background: Heritability is a central measure in genetics quantifying how much of the variability observed in a trait is attributable to genetic differences. Existing methods for estimating heritability are most often based on random-effect models, typically for computational reasons. The alternative of using a fixed-effect model has received much more limited attention in the literature. Results: In this paper, we propose a generic strategy for heritability inference, termed as boosting heritability, by combining several advantageous features of different recent methods to produce an estimate of the heritability with a high-dimensional linear model. Boosting heritability uses in particular a multiple sample splitting strategy which leads to a more reliable estimate. We use both simulated data and real antibiotic resistance data from a major human pathogen, Sptreptococcus pneumoniae, to demonstrate the attractive features of our inference strategy. Conclusions: Compared with other methods, boosting heritability yields a more reliable estimate and allows one to make inference about heritability.

Sign in / Sign up

Export Citation Format

Share Document