• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.252-265
• /
• 2022

Subsurface velocity is critical for the accurate resolution geological structures. The estimation of acoustic impedance is also critical, as it provides key information regarding the reservoir properties. Therefore, researchers have developed various inversion approaches for the estimation of reservoir properties. The Markov chain Monte Carlo method, which is a stochastic method, has advantages over the deterministic method, as the stochastic method enables us to attenuate the local minima problem and quantify the uncertainty of inversion results. Therefore, the Markov chain Monte Carlo inversion method has been applied to various kinds of geophysical inversion problems. However, studies on the Markov chain Monte Carlo inversion are still very few compared with deterministic approaches. In this study, we reviewed various types of reversible jump Markov chain Monte Carlo applications and explained the key concept of each application. Furthermore, we discussed future applications of the stochastic method.

• The Korean Journal of Applied Statistics
• /
• v.26 no.4
• /
• pp.649-659
• /
• 2013

We use a Markov chain model to analyze the Korean Baseball League. We derive the distributions of the number of runs scored and the number of batters that complete their turn at bat in a baseball game using the time inhomogeneous Markov chain. The model is tested with real data produced from the 2011 Korean Baseball League.

• KIISE Transactions on Computing Practices
• /
• v.21 no.1
• /
• pp.94-99
• /
• 2015

With the recent machine learning paradigm of using nonparametric Bayesian statistics and statistical inference based on random sampling, the Dirichlet distribution finds many uses in a variety of graphical models. It is a multivariate generalization of the gamma distribution and is defined on a continuous (K-1)-simplex. This paper presents a sampling method for a Dirichlet distribution for the problem of dividing an integer X into a sequence of K integers which sum to X. The target samples in our problem are all positive integer vectors when multiplied by a given X. They must be sampled from the correspondingly gridded simplex. In this paper we develop a Markov Chain Monte Carlo (MCMC) proposal distribution for the neighborhood grid points on the simplex and then present the complete algorithm based on the Metropolis-Hastings algorithm. The proposed algorithm can be used for the Markov model, HMM, and Semi-Markov model for accurate state-duration modeling. It can also be used for the Gamma-Dirichlet HMM to model q the global-local duration distributions.

• 한국해양학회지
• /
• v.22 no.2
• /
• pp.57-62
• /
• 1987

The Markov chain properties of the sea surface temperature (SST) anomalies, namely, the dependency of the monthly SST anomaly on that of the previous month, are studied based on the SST data for 28years(1957-1984) at 5 stations in the southeastern coast of Korea. Wi classified the monthly SST anomalies at each station into the low, the normal and the high state, and computed transition probabilities between SST anomalies of two successive months The standard deviation of SST anomalies at each station is used as a reference for the classification of SST anomalies into 3states. The transition probability of the normal state to remain in the same state is about 0.8. The transition probability of the high or the low states to remain in the same state is about one half. The SST anomalies have almost no probability to transit from the high (the low) state to the low (the high) state. Statistical tests show that the Markov chain properties of SST anomalies are stationary in tine and homogeneous in space. The multi-step Markov chain analysis shows that the 'memory' of the SST anomalies at the coastal stations remains about 3 months.

• Proceedings of the Korea Institutes of Information Security and Cryptology Conference
• /
• 1991.11a
• /
• pp.105-133
• /
• 1991

본 논문에서는 임의의 이진 난수발생기의 source가 $BMS_{p}$ 이거나 M-memory를 갖는 마르코프연쇄로 모델화 되었을 경우에 비트당 entropy와 관련이 있는 새로운 randomness에 관한 통계적 검정법을 제안한다. 기존에 알려진 이진 난수발생기의 randomness검정법이 0또는 1의 분포의 편향성(bias)이나 연속된 비트간의 상관성(correlation)중의 한 종류만의 non-randomness를 추적해낼 수 있는 반면에 새로운 검정법은 위의 두가지 검정을 통과하였을 때 암호학적으로 중요한 측도인 비트당 entropy 를 측정하여 암호학적인 약점을 검정할 수 있다. 또한 대칭(비밀키) 암호시스템의 통계적 결점을 바탕으로 하여 키를 찾는 공격자의 최적 전략( optimal strategy)문제를 분석하여 이 최적 전략이 이진 수열의 비트당 entropy와 밀접한 관계가 있음을 보이고 이 비트당 entropy와 관련이 있는 새로운 통계량을 도입하여 이진 난수 발생기의 source의 이진수열이 다음 3가지 경우, 즉, i.i.d. symmetric인 경우, $BMS_{p}$ 인 경우, M-memory를 갖는 마르코프연쇄인 경우의 각각에 대하여 특성을 조사하고 새로운 통계량의 평균과 분산을 구한다. 이때 구한 새로운 통계량은 잘 알려진 중심 극한 정리에 의하여 근사적으로 정규분포를 따르므로 위의 평균과 분산을 이용하여 스트림 암호시스템에서 구성요소로 많이 사용되는 몇 몇 간단한 이진 난수 발생기에 적용하여 통계적 검정을 실시함으로써 entropy 관점의 검정법이 새로운 randomness 검정법으로 타당함을 보인다.

• Journal of KIISE
• /
• v.42 no.6
• /
• pp.748-754
• /
• 2015

This paper proposes a sequential state estimation model consisting of continuous and discrete variables, as a way of generalizing all discrete-state factorial HMM, and gives a design of gait motion model based on the idea. The discrete state variable implements a Markov chain that models the gait dynamics, and for each state of the Markov chain, we created a Gaussian process over the space of the continuous variable. The Markov chain controls the switching among Gaussian processes, each of which models the rotation or various views of a gait state. Then a particle filter-based algorithm is presented to give an approximate filtering solution. Given an input vector sequence presented over time, this finds a trajectory that follows a Gaussian process and occasionally switches to another dynamically. Experimental results show that the proposed model can provide a very intuitive interpretation of video-based gait into a sequence of poses and a sequence of posture states.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.2
• /
• pp.411-423
• /
• 2009

In the current paper, by extending Verall (1990)'s work, we propose a new Bayesian model for analyzing run-off triangle data. While Verall's (1990) work only account for the calendar year and evolvement time effects, our model further accounts for the "absolute time" effects. We also suggest a Markov Chain Monte Carlo method that can be used for estimating the proposed model. We apply our proposed method to analyzing three empirical examples. The results demonstrate that our method significantly reduces prediction error when compared with the existing methods.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.3
• /
• pp.240-246
• /
• 2015

In this paper, speech enhancement using nonnegative matrix factorization with temporal continuity has been addressed. Speech and noise signals are modeled as Possion distributions, and basis vectors and gain vectors of NMF are modeled as Gamma distributions. Temporal continuity of the gain vector is known to be critical to the quality of enhanced speech signals. In this paper, temporal continiuty is implemented by adopting Gamma-Markov chain priors for noise gain vectors during the separation phase. Simulation results show that the Gamma-Markov chain models temporal continuity of noise signals and track changes in noise effectively.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.17 no.4
• /
• pp.331-337
• /
• 2011

We described for the production of cloud-free satellite sea surface temperature(SST) data around Northeast Asian using NOAA AVHRR(Advanced Very High Resolution Radiometer) SST data during 1990-2005. As a result of Markov model, it was found that the value of Markov coefficient in the strong current region such as Kuroshio region showed smaller than that in the weak current. The variations of average SST and regional difference of seasonal day-to-day SST in spring and fall were larger than those in summer and winter. In particular, the distribution of the regional difference appeared large in the vicinity of continental in spring and fall. The difference of seasonal day-to-day SST was also small in Kuroshio region and southern part of East Sea due to the heat advection by warm currents.

• The Korean Journal of Applied Statistics
• /
• v.29 no.2
• /
• pp.321-330
• /
• 2016

Run expectancy (RE) is the mean number of runs scored from a specific base runner/outs situation of an inning to the end of the inning. Win expectancy (WE) is the probability that a particular team will win the game at a specific game state such as half-inning, score difference, outs, and/or runners on base. In this paper, we derive RE and WE for the Korea Baseball Organization (KBO) League based on six-year data from 2007 to 2012 using a Markov chain model.