• 한국품질경영학회:학술대회논문집
• /
• 한국품질경영학회 1998년도 The 12th Asia Quality Management Symposium* Total Quality Management for Restoring Competitiveness
• /
• pp.225-232
• /
• 1998

A Markov Chain Monte Carlo method with data augmentation is developed to compute the features of the posterior distribution. This data augmentation approach facilitates the specification of the transitional measure in the Markov Chain. Bayesian analysis of the mixture exponential model discusses using the Gibbs sampler. Parameter and reliability estimators are obtained. A numerical study is provided.

• 대한수학회논문집
• /
• 제14권3호
• /
• pp.627-633
• /
• 1999

Let the given distribution $\pi$ have a log-concave density which is proportional to exp(-V(x)) on $R^d$. We consider a Markov chain induced by the method Gibbs sampling having $\pi$ as its in-variant distribution and prove geometric ergodicity and the functional central limit theorem for the process.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2003년도 춘계 학술발표회 논문집
• /
• pp.183-189
• /
• 2003

A momentum threshold autoregressive (MTAR) model, a nonlinear autoregressive model, is analyzed in a Bayesian framework. Parameter estimation in the presence of missing data is done by using Markov chain Monte Carlo methods. We also propose simple Bayesian test procedures for asymmetry and unit roots. The proposed method is applied to a set of Korea unemployment rate data and reveals evidence for asymmetry and a unit root.

• Communications for Statistical Applications and Methods
• /
• 제9권1호
• /
• pp.115-128
• /
• 2002

Hierarchical models are widely used for inference on correlated parameters as a compromise between underfitting and overfilling problems. In this paper, we take a Bayesian approach to analyzing hierarchical models and suggest a Markov chain Monte Carlo methods to get around computational difficulties in Bayesian analysis of the hierarchical models. We apply the method to a real data on smoking and lung cancer which are collected from cities in China.

• Smart Structures and Systems
• /
• 제15권3호
• /
• pp.735-749
• /
• 2015

The major difficulty of using Bayesian probabilistic inference for system identification is to obtain the posterior probability density of parameters conditioned by the measured response. The posterior density of structural parameters indicates how plausible each model is when considering the uncertainty of prediction errors. The Markov chain Monte Carlo (MCMC) method is a widespread medium for posterior inference but its convergence is often slow. The differential evolution adaptive Metropolis-Hasting (DREAM) algorithm boasts a population-based mechanism, which nms multiple different Markov chains simultaneously, and a global optimum exploration ability. This paper proposes an improved differential evolution adaptive Metropolis-Hasting algorithm (IDREAM) strategy to estimate the posterior density of structural parameters. The main benefit of IDREAM is its efficient MCMC simulation through its use of the adaptive Metropolis (AM) method with a mutation strategy for ensuring quick convergence and robust solutions. Its effectiveness was demonstrated in simulations on identifying the structural parameters with limited output data and noise polluted measurements.

• International Journal of Contents
• /
• 제13권4호
• /
• pp.80-87
• /
• 2017

Since the advent of the Internet, and the development of smart devices, people have begun to spend more time in online platforms; this phenomenon has created a large number of online Words of Mouth (WOM) daily. Under these changes, one of the important aspects to consider is the conformity effect in online WOM; that is, whether an individual's own opinion would be influenced by the majority opinion of other people. This study, therefore, investigates whether there is the conformity effect in online product ratings for Amazon.com using the method called Markov Chain analysis. Markov Chain analysis considers the stochastic process that satisfies the Markov property, and we assume that the generation of online product ratings follows the process. Under the assumption that people are usually independent when they express their opinion in online platforms, we analyze the interdependency among rating sequences, and we find weak evidence that there exists the conformity effect in online product rating. This suggests that people who leave online product ratings consider others' opinions.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2008년도 학술대회 1부
• /
• pp.739-743
• /
• 2008

본 연구에서는 사용자가 입력한 멜로디에 따른 반주 음악을 자동으로 생성하는 방법을 제시한다. 시작되는 코드는 사용자의 멜로디에 의해서 생성이 되며, 그 다음 코드들은 코드들간의 전이확률이 정의되어있는 마르코프 체인(markov chain)의 확률 테이블을 이용하여 연속적으로 생성된다. 확률 테이블은 기존 음악의 샘플 데이터를 강화학습(reinforcement learning)을 이용하여 학습된다. 또한 실시간으로 재생되는 반주 코드는 매 상태 마다 주어지는 보상 값을 통해 더 나은 행동을 취할 수 있도록 학습해 나간다. 멜로디와 각 코드들간의 유사성은 피치 클래스 히스토그램을 이용하여 계산된다. 본 기술을 사용하여 주어진 사용자 입력에 조화로운 반주 코드의 자동 생성이 가능하다.

• Structural Engineering and Mechanics
• /
• 제63권1호
• /
• pp.47-53
• /
• 2017

The fundamental goal of this study is to minimize the uncertainty of the median fragility curve and to assess the structural vulnerability under earthquake excitation. Bayesian Inference with Markov Chain Monte Carlo (MCMC) simulation has been presented for efficient collapse response assessment of the independent intake water tower. The intake tower is significantly used as a diversion type of the hydropower station for maintaining power plant, reservoir and spillway tunnel. Therefore, the seismic fragility assessment of the intake tower is a pivotal component for estimating total system risk of the reservoir. In this investigation, an asymmetrical independent slender reinforced concrete structure is considered. The Bayesian Inference method provides the flexibility to integrate the prior information of collapse response data with the numerical analysis results. The preliminary information of risk data can be obtained from various sources like experiments, existing studies, and simplified linear dynamic analysis or nonlinear static analysis. The conventional lognormal model is used for plotting the fragility curve using the data from time history simulation and nonlinear static pushover analysis respectively. The Bayesian Inference approach is applied for integrating the data from both analyses with the help of MCMC simulation. The method achieves meaningful improvement of uncertainty associated with the fragility curve, and provides significant statistical and computational efficiency.

• The Journal of Asian Finance, Economics and Business
• /
• 제7권8호
• /
• pp.281-287
• /
• 2020

In capital budgeting practices, investment project evaluations based on the net present value (NPV) and the internal rate of return (IRR) represent the traditional evaluation techniques. Compared with the traditional methods, the modified internal rate of return (MIRR) gives the opportunity to evaluate an investment in certain projet, while taking the changes in cash flows over time and issuing shares such as dividing shares, bonuses, and dividend for each end of the investment year into account. Therefore, this study aims to evaluate an investment in the Malaysian construction sector utilizing financial data for 39 public listed companies operating in the Malaysian construction sector over the period from Jan 1, 2007, to December 30, 2018, based on the MIRR method. Stochastic was studied in this study to estimate the estimated probability by applying the Markov chain model to the MIRR method where the transition matrix has two possible movements of either Good (G) or Bad (B). it is found that the long-run probability of getting a good investment is higher than the probability of getting a bad investment in the long-run, where were the probabilities of good and bad are 0.5119, 0.4881, respectively. Hence, investment in the Malaysian construction sector is recommended.

• 전기학회논문지P
• /
• 제58권2호
• /
• pp.164-171
• /
• 2009

This paper presents new learning algorithm of dynamic Bayesian networks (DBN) by means of constrained least square (LS) estimation algorithm and gradient descent method. First, we propose constrained LS based parameter estimation for a Markov chain (MC) model given observation data sets. Next, a gradient descent optimization is utilized for online estimation of a hidden Markov model (HMM), which is bi-linearly constructed by adding an observation variable to a MC model. We achieve numerical simulations to prove its reliability and superiority in which a series of non stationary random signal is applied for the DBN models respectively.