Comparing the MCMC Efficiency of JAGS and Stan for the Multi-Level Intercept-Only Model in the Covariance- and Mean-Based and Classic Parametrization

Bayesian MCMC is a widely used model estimation technique, and software from the BUGS family, such as JAGS, have been popular for over two decades. Recently, Stan entered the market with promises of higher efficiency fueled by advanced and more sophisticated algorithms. With this study, we want to contribute empirical results to the discussion about the sampling efficiency of JAGS and Stan. We conducted three simulation studies in which we varied the number of warmup iterations, the prior informativeness, and sample sizes and employed the multi-level intercept-only model in the covariance- and mean-based and in the classic parametrization. The target outcome was MCMC efficiency measured as effective sample size per second (ESS/s). Based on our specific (and limited) study setup, we found that (1) MCMC efficiency is much higher for the covariance- and mean-based parametrization than for the classic parametrization, (2) Stan clearly outperforms JAGS when the covariance- and mean-based parametrization is used, and that (3) JAGS clearly outperforms Stan when the classic parametrization is used.

Bayesian Long-Short Term Memory for History Matching in Reservoir Simulations

History matching is critical in subsurface flow modeling. It is to align the reservoir model with the measured data. However, it remains challenging since the solution is not unique and the implementation is expensive. The traditional approach relies on trial and error, which are exhaustive and labor-intensive. In this study, we propose a new workflow utilizing Bayesian Markov Chain Monte Carlo (MCMC) to automatically and accurately perform history matching. We deliver four novelties within the workflow: 1) the use of multi-resolution low-fidelity models to guarantee high-quality matching, 2) updating the ranges of priors to assure convergence, 3) the use of Long-Short Term Memory (LSTM) network as a low-fidelity model to produce continuous time-response, and 4) the use of Bayesian optimization to obtain the optimum low-fidelity model for Bayesian MCMC runs. We utilize the first SPE comparative model as the physical and high-fidelity model. It is a gas injection into an oil reservoir case, which is the gravity-dominated process. The coarse low-fidelity model manages to provide updated priors that increase the precision of Bayesian MCMC. The Bayesian-optimized LSTM has successfully captured the physics in the high-fidelity model. The Bayesian-LSTM MCMC produces an accurate prediction with narrow uncertainties. The posterior prediction through the high-fidelity model ensures the robustness and precision of the workflow. This approach provides an efficient and high-quality history matching for subsurface flow modeling.

Estimasi Jumlah Penghuni Ruangan Berdasarkan Konsentrasi CO2 Dengan Metode Bayesian MCMC

<span lang="PT-BR">The number of occupants in the building is important information for building management because it is related to security issues, evacuation, and energy saving. This article focuses on estimating the number of occupants using the Bayesian Monte Carlo Markov chain (MCMC) method based on indoor CO₂ levels. Probability theory underlies the Bayesian MCMC principle, where the mass balance equation of indoor CO₂ is used as a physical model of estimation calculations. Determination of the variables in the mass balance equation is investigated to obtain the effect on the accuracy of the estimated number of occupants. It found that the higher the standard deviation of the input variable on the physical model, the higher the error estimation produced. In addition, the Bayesian MCMC algorithm is tested in a real-time scheme of test</span><span lang="IN">-</span><span lang="PT-BR">chamber. The result shows an estimated error of 39%. Rapid changes influence estimation errors in actual occupants relative to the sample interval and the time delay of the estimation.</span>

Are Entrepreneurs More Likely to Be Obese?

Background: Entrepreneurs not only promote a nation’s economic growth but also increase employment. The risk of obesity among entrepreneurs may bring heavy economic burdens not only to the entrepreneurs but also to the national health care system. We aimed to examine the association between entrepreneurship and the risk of obesity. Methods: We utilized data from the 2015 Harmonized China Health and Retirement Longitudinal Survey, including 2,802 individuals aged between 45 and 65 with complete data. This study used BMI (Body Mass Index) (kg/m2 ) as an indicator of obesity risk. Entrepreneurs were defined as those respondents who run their own businesses as main jobs. We used multivariate OLS regression models and Bayesian Markov Chain Monte Carlo (MCMC) method to examine the link of entrepreneurship and obesity risk. Results: The multivariate OLS regression results showed that entrepreneurship was positively associated with BMI (P<0.01). The Bayesian MCMC results indicated that the posterior mean was (0.597, 90% HPD CI: 0.319, 0.897), demonstrating that entrepreneurship was indeed significantly positively associated with the risk of obesity. Conclusion: Being an entrepreneur is positively associated with the risk of obesity. As obesity can cause diseases such as hypertension, diabetes, coronary heart disease and stroke, the health departments should take necessary health interventions to prevent entrepreneurs from being obese in order to increase their entrepreneurial success.

Mechanical Ventilation Control Based on Estimated occupancy using a Carbon Dioxide Sensor

Ventilation is an important aspect to maintain good indoor air quality in a building. However, excessive ventilation causing high energy consumption of the HVAC system. The ASHRAE Standard provides a guideline to set the ventilation rate that depends on the occupants' number and space. Thus, quantification of the number of occupants is required to regulate the ventilation rate. In this study, the estimated number of occupants was estimated using a Bayesian MCMC method based on CO2 levels. The mass balance equation of the CO2 is used as a model for the calculation of Bayesian MCMC. The Bayesian method for estimating the occupants' number is tested in a 96,7 m3 office room equipped with a ventilation system. Thus the occupancy estimation and control of ventilation can be done in real-time. The test also includes conventional ventilation control based on CO2 levels directly without converting to the occupants' number. The ventilation rate based on the number of occupants at the present test chamber refers to ASHRAE 62.1. The test results show that ventilation controlled by the estimated number of occupants using the Bayesian method successfully conducted with ventilation rate per occupant closer to the ASHRAE 62.1 standard over conventional ventilation method

BAYESIAN REGRESI SURVIVAL PADA PROSES KEJADIAN HIV/AIDS DI JAWA TIMUR

Survival analysis is a statistical procedure that describes a mathematical model that is often applied in various studies, especially in health. One application of survival analysis is to determine the rate of survival and the factors affecting HIV / AIDS sufferers in East Java. HIV / AIDS is a virus that attacks or infects white blood cells, causing a decrease in immune cells. This disease causes a decrease in productivity in the health and economic sectors of a country. Even if the disease continues to increase, the weak economic development will decrease due to the treatment of HIV/AIDS and the risk of death of people infected with the HIV / AIDS virus is getting higher in East Java. In addition to these health and economic quality factors, factors such as residents' knowledge of the disease. By knowing the factors of HIV/AIDS survival rate, mathematical modelling can be done to estimate the duration of the patient's survival power comprehensively and accurately. In this study, we want to find out what factors affect the survival rate of HIV/AIDS using the 3-Parameter Lognormal Survival Link Function model in which the method of parameter estimation used is the Bayesian MCMC-Gibbs Sampling method. The best models is the 3-parameter lognormal survival with frailty that is normally distributed and factors affect the survival rate of HIV/AIDS is education (X3), marital status (X5), Stadium of the patient (X8), adherence of therapy (X10), opportunistic infection (X11) and risk factor of infection (X13). Analisis survival merupakan suatu prosedur statistika yang menjelaskan model matematis yang seringkali diaplikasikan dalam berbagai penelitian, terutama di bidang kesehatan. Salah satu penerapan dari analisis survival adalah untuk mengetahui laju bertahan hidup dan faktor-faktor yang mempengaruhi penderita HIV/AIDS di Jawa Timur. Penyakit HIV/AIDS adalah virus yang menyerang atau menginfeksi sel darah putih yang menyebabkan turunnya sel kekebalan tubuh. Penyakit ini mengakibatkan penurunan produktivitas di bidang kesehatan dan ekonomi di suatu negara. Bahkan apabila penyakit ini terus meningkat maka lemahnya perkembangan ekonomi akan menjadi menurun akibat pengobatan penyakit HIV/AIDS dan resiko kematian dari orang yang terinfeksi virus HIV/AIDS tersebut semakin tinggi di Jawa Timur. Disamping faktor kualitas kesehatan dan ekonomi tersebut, faktor seperti pengetahuan warga terhadap penyakit HIV/AIDS. Dengan mengetahui faktor-faktor laju bertahan hidup penyakit HIV/AIDS dapat dilakukan pemodelan matematis untuk memperkirakan durasi daya survival secara aktual, dan komprehensif. Tujuan artikel dalam penelitian ini adalah menjelaskan faktor-faktor yang mempengaruhi laju bertahan hidup pasien terhadap penyakit HIV/AIDS dengan menggunakan model Survival Lognormal 3 parameter Link Function. Metode estimasi parameter yang digunakan adalah metode Bayesian MCMC-Gibbs Sampling. Model Survival Lognormal 3 Parameter dengan Frailty yang berdistribusi normal menghasilkan faktor-faktor yang mempengaruhi laju bertahan hidup pasien HIV/AIDS di Jawa Timur adalah pendidikan(X3), status perkawinan (X5), stadium penderita (X8), kepatuhan terapi (X10), infeksi oportunitis (X11) dan resiko penularan (X13).

Bayesian Estimation of the Functional Spatial Lag Model

The spatial lag model (SLM) has been widely studied in the literature for spatialised data modeling in various disciplines such as geography, economics, demography, regional sciences, etc. This is an extension of the classical linear model that takes into account the proximity of spatial units in modeling. In this paper, we propose a Bayesian estimation of the functional spatial lag (FSLM) model. The Bayesian MCMC technique is used as a method of estimation for the parameters of the model. A simulation study is conducted in order to compare the results of the Bayesian functional spatial lag model with the functional spatial lag model and the functional linear model. As an illustration, the proposed Bayesian functional spatial lag model is used to establish a relationship between the unemployment rate and the curves of illiteracy rate observed in the 45 departments of Senegal.