Bayesian Ordered Probit Model Research Articles

Opposite Effects of Work-Related Physical Activity and Leisure-Time Physical Activity on the Risk of Diabetes in Korean Adults.

The object of this study was to examine the effects of domestic and work-related physical activity (DWPA) and leisure-time physical activity (LTPA) on the risk of diabetes, by categorizing fasting blood glucose (FBG) levels into normal, Impaired Fasting Glucose (IFG), and diabetes. The sample consisted of 4661 adults aged 30 years or above, and was chosen from the 2017 Korean National Health and Nutrition Examination Survey (KNHANES) data. Of all the subjects, 14.6% engaged in high-intensity DWPA and 6.25% in moderate-intensity DWPA; while 11.68% and 24.80% engaged in high- and moderate-intensity LTPA, respectively. The effects of both types of physical activities on the risk of diabetes were analyzed using a Bayesian ordered probit model. For those with high-intensity DWPA, the probability of the FBG level being normal was 5.10% (SE = 0.25) lower than for those with non-high-intensity DWPA, and the probabilities of IFG and diabetes were 3.30% (SE = 0.15) and 1.79% (SE = 0.09) higher, respectively. However, for those with high-intensity LTPA, the probability of the FBG level being normal was 2.54% (SE = 0.09) higher, and the probabilities of IFG and diabetes were 1.74% (SE = 0.07) and 0.80% (SE = 0.03) lower, respectively, than those with non-high-intensity LTPA. Likewise, for moderate-intensity DWPA and LTPA, the results were the same compared to low-intensity physical activities though the magnitude of the effects were smaller than for high-intensity. Thus, the activities related to work have a negative effect and those related to leisure have a positive effect. The criteria for physical activities to reduce the risk of diabetes should be set by separating these domains of physical activity, and new management strategies for diabetes are needed for people with moderate- or high-intensity DWPA.

Genome-based prediction of Bayesian linear and non-linear regression models for ordinal data.

Linear and non-linear models used in applications of genomic selection (GS) can fit different types of responses (e.g., continuous, ordinal, binary). In recent years, several genomic-enabled prediction models have been developed for predicting complex traits in genomic-assisted animal and plant breeding. These models include linear, non-linear and non-parametric models, mostly for continuous responses and less frequently for categorical responses. Several linear and non-linear models are special cases of a more general family of statistical models known as artificial neural networks, which provide better prediction ability than other models. In this paper, we propose a Bayesian Regularized Neural Network (BRNNO) for modelling ordinal data. The proposed model was fitted using a Bayesian framework; we used the data augmentation algorithm to facilitate computations. The proposed model was fitted using the Gibbs Maximum a Posteriori and Generalized EM algorithm implemented by combining code written in C and R programming languages. The new model was tested with two real maize datasets evaluated for Septoria and GLS diseases and was compared with the Bayesian Ordered Probit Model (BOPM). Results indicated that the BRNNO model performed better in terms of genomic-based prediction than the BOPM model.

Probabilistic performance prediction of shield tunnels in operation through data mining

To investigate the inherent uncertain and dynamic deterioration of metro shield tunnels in operation, the Bayesian ordered probit model, a data mining method, was used in this study. Through Markov Chain Monte Carlo (MCMC) simulation, the uncertainty in parameter estimation was significantly reduced, and the confidence in the results was improved. The effects of influencing variables on the deterioration process were evaluated. It was found that the tunnel sections with greater burial depths were more likely to deteriorate than the shallow ones. Crossing below a river or near a station or cross passage accelerated the deterioration rate. The deterioration probability increased as the tunnel became older. Finally, the model was applied to a tunnel section. It was shown that the probability of the best state decreased while that of the worst state increased with age. For states between the best and the worst, the probability would first increase, reach a peak, and then decrease. This study found that the ordered probit model with MCMC was a valuable probabilistic method for performance prediction, which is crucial for cost-effective decision-making in future work.

Understanding Sovereign Rating Movements in Euro Area Countries

This paper investigates the link between sovereign ratings and macroeconomic fundamentals for a group of euro area countries which recorded rating downgrades amid the euro area sovereign debt crisis. We apply an elaborated econometric estimation technique, based on a Bayesian ordered probit model, to understand how the decisions of rating agencies can be explained by economic developments. The estimated model re-produces historical ratings by using a small number of economic and institutional variables, which seem to effectively summarize the large number of criteria used by Moody JEL Classification: C25, G24, H63, H68

Comparison of Waist-to-height Ratio (WHtR), Body Mass Index (BMI) and Waist Circumference (WC) as a Screening Tool for Prediction of Metabolic-related Diseases

The present study showed WHtR to be significantly better than BMI and WC for prediction of metabolic-related diseases in the middle-aged and older people in Korea, based on Bayesian ordered probit model analysis. The variations of WC, BMI and WHtR were compared according to the number of metabolic-related diseases such as hypertension, dyslipidemia, stroke, myocardial infarction, angina pectoris and diabetes. It was found that the three measures showed the similar variation except a very few extreme cases for age less than 40. For subjects over the age of 40, WC was not significant and WHtR gave more influence in greater variability than BMI on the number of metabolic diseases. Also, the rate of change for WHtR was higher than for BMI as the number of metabolic-related diseases increased. Specifically, the difference of the marginal effect of WHtR between no disease and only one disease was 1.81 times higher than that of BMI. Moreover, it was pointed out that the threshold value of WHtR for obesity should be considered differently by age.

Two-Vehicle Injury Severity Models Based on Integration of Pavement Management and Traffic Engineering Factors

Objective: The severity of traffic-related injuries has been studied by many researchers in recent decades. However, the evaluation of many factors is still in dispute and, until this point, few studies have taken into account pavement management factors as points of interest. The objective of this article is to evaluate the combined influences of pavement management factors and traditional traffic engineering factors on the injury severity of 2-vehicle crashes. Methods: This study examines 2-vehicle rear-end, sideswipe, and angle collisions that occurred on Tennessee state routes from 2004 to 2008. Both the traditional ordered probit (OP) model and Bayesian ordered probit (BOP) model with weak informative prior were fitted for each collision type. The performances of these models were evaluated based on the parameter estimates and deviances. Results: The results indicated that pavement management factors played identical roles in all 3 collision types. Pavement serviceability produces significant positive effects on the severity of injuries. The pavement distress index (PDI), rutting depth (RD), and rutting depth difference between right and left wheels (RD_df) were not significant in any of these 3 collision types. The effects of traffic engineering factors varied across collision types, except that a few were consistently significant in all 3 collision types, such as annual average daily traffic (AADT), rural–urban location, speed limit, peaking hour, and light condition. Conclusions: The findings of this study indicated that improved pavement quality does not necessarily lessen the severity of injuries when a 2-vehicle crash occurs. The effects of traffic engineering factors are not universal but vary by the type of crash. The study also found that the BOP model with a weak informative prior can be used as an alternative but was not superior to the traditional OP model in terms of overall performance.

The Organization of Political Violence by Insurgencies

AbstractA range of theories have attempted to explain the variation in civilian abuse of warring parties. Most of these theories have been focused on the strategic environment in which these acts take place. Less attention is devoted to the perpetrators of these human right abuses themselves: the armed groups. This study tries to fill this niche by using the organizational process theory in which it is assumed that armed groups, like every organization, struggles for survival. The leader tries to ensure the maintenance of her armed group by increasing her control over her troops. The relationship between the level of control and the perpetrated civilian abuse is examined with a new dataset on the internal structure of more than 70 different armed groups around the world. With the help of a Bayesian Ordered Probit model, this new dataset on civilian abuse is analyzed. The results show that especially particular incentives play an important role.

Measuring inequality of subjective well-being: A Bayesian approach

Abstract This article proposes a new measure for inequality of subjective well-being and shows the empirical results using a Bayesian ordered probit model. We introduce a new concept called “regret” as a measure for inequality of subjective well-being. Regret is the probability with which a respondent who chooses an option in a multiple-choice question pertaining to subjective well-being does not choose any other option indicative of better well-being. Regret is estimated in connection with demographic factors using the Markov chain Monte Carlo (MCMC) method and data of the World Values Survey. Furthermore, the relationships between regret and GDP per capita and the changes therein are shown to investigate those between inequality of subjective well-being and economic conditions.

Crash Injury Severity Analysis Using Bayesian Ordered Probit Models

Understanding the underlying relationship between crash injury severity and factors such as driver's characteristics, vehicle type, and roadway conditions is very important for improving traffic safety. Most previous studies on this topic used traditional statistical models such as ordered probit (OP), multinomial logit, and nested logit models. This research introduces the Bayesian inference and investigates the application of a Bayesian ordered probit (BOP) model in driver's injury severity analysis. The OP and BOP models are compared based on datasets with different sample sizes from the 2003 National Automotive Sampling System General Estimates System (NASSGES). The comparison results show that these two types of models produce similar results for large sample data. When the sample data size is small, with proper prior setting, the BOP model can produce more reasonable parameter estimations and better prediction performance than the OP model. This research also shows that the BOP model provides a flexible framework that can combine information contained in the data with the prior knowledge of the parameters to improve model performance.

A Quantitative Analysis on Tourists' Consumer Satisfaction via the Bayesian Ordered Probit Model

本研究では，ベイズ順序プロビットモデルを用いて観光客の顧客満足度(CS)を分析することを目的としている．鎌倉市での観光CS調査データを用いて，観光客の個人属性，観光行動と観光CSとの定量的，実証的分析を行った．モデルパラメータは，マルコフチェイン・モンテカルロ法の代表的手法であるギブス・サンプリング法を用いて推定した．限界効果分析の結果，観光客の観光全般の満足度が，旅行目的，立ち寄り地点，個人属性などに強く影響を受けることが，定量的に示された．MCMC法によるベイズ順序プロビットモデルは，最尤推定法で解が求められない場合や満足度の予測に有効であることが示された．今後，来訪頻度やスケジューリングなどの観光行動分析などへの応用が課題である．