• Journal of the Korean Data and Information Science Society
• /
• 제8권2호
• /
• pp.246-246
• /
• 1997

• Communications for Statistical Applications and Methods
• /
• 제19권3호
• /
• pp.495-500
• /
• 2012

We consider a Bayesian test of independence in a two-way contingency table that has some zero cells. To do this, we take a three-stage hierarchical Bayesian model under each hypothesis. For prior, we use Dirichlet density to model the marginal cell and each cell probabilities. Our method does not require complicated computation such as a Metropolis-Hastings algorithm to draw samples from each posterior density of parameters. We draw samples using a Gibbs sampler with a grid method. For complicated posterior formulas, we apply the Monte-Carlo integration and the sampling important resampling algorithm. We compare the values of the Bayes factor with the results of a chi-square test and the likelihood ratio test.

• 대한치과의사협회지
• /
• 제54권11호
• /
• pp.850-864
• /
• 2016

The Generalized Estimating Equations (GEE) approach is a widely used statistical method for analyzing longitudinal data and clustered data in clinical studies. In dentistry, due to multiple outcomes obtained from one patient, the outcomes produced from an individual patient are correlated with one another. This study focused on the basic ideas of GEE and introduced the types of covariance matrix and working correlation matrix. The quasi-likelihood information criterion (QIC) and quasi-likelihood information criterion approximation ($QIC_u$) were used to select the best working correlation matrix and the best fitting model for the correlated outcomes. The purpose of this study is to show a detailed process for the GEE analysis using SPSS software along with an orthodontic miniscrew example, and to help understand how to use GEE analysis in dental research.

• 응용통계연구
• /
• 제6권1호
• /
• pp.41-52
• /
• 1993

본 논문에서는 과산포 혼합 모형인 음이항 분포와 베타이항 분포에서 피어슨 형태 및 데비 언스 형태의 분산치 교정에 대한 효과를 수리적으로 비교했다. 이들 과산포 혼합 모형은, 평 균과 분산을 동시에 모형화 하는데 매우 유용한 준우도함수의 중요한 구성원이다. 모의실험 을 통해서 분산치의 교정이 평균, 산포모수에 따라 어떻게 달라지는지 비교 연구하였다.

• 한국통신학회논문지
• /
• 제32권12A호
• /
• pp.1219-1225
• /
• 2007

본 논문은 4개의 송신 안테나를 사용 할 경우 임의의 QAM 변조 방식과 준정지 Rayleigh 페이딩 채널 환경에서 최대 다이버시티 이득을 획득하고 더불어 full-rate을 획득하는 새로운 QO-STBC(quasi-orthogonal space-time block code)를 제안한다. 특히 제안된 부호는 기존의 MDC-QO-STBC(minimum decoding complex QO-STBC)와 같이 준직교 특성을 만족하여 수신단에서 2개의 실수 심벌들로 구성된 4개의 그룹들에 대하여 각각 독립적인 ML(maximum likelihood) 복호가 가능하다는 장점을 갖는다. 전산 실험 결과 제안된 부호는 모든 변조 방식들에 대하여 MDC-QO-STBC와 동일한 비트 오율(BER) 성능을 보여준다. 하지만 제안된 부호는 모든 송신 안테나들에서 성상 회전 이전의 신호들에 비해 PAPR(peak-to-average power ratio) 값들이 증가하는 기존의 MDC-QO-STBC와 달리 오직 2개의 송신 안테나들에서만 PAPR의 증가가 발생하도록 수정 가능하며 따라서 실제 송신 시스템 구현 시 훨씬 큰 이점을 갖게 된다.

• Communications for Statistical Applications and Methods
• /
• 제19권6호
• /
• pp.761-770
• /
• 2012

We frequently encounter outcomes of count that have extra variation. This paper considers several alternative models for overdispersed count responses such as a quasi-Poisson model, zero-inflated Poisson model and a negative binomial model with a special focus on a generalized linear mixed model. We also explain various goodness-of-fit criteria by discussing their appropriateness of applicability and cautions on misuses according to the patterns of response categories. The overdispersion models for counts data have been explained through two examples with different response patterns.

• 한국통신학회논문지
• /
• 제23권6호
• /
• pp.1583-1591
• /
• 1998

이 논문에서는 시간을 따라 바뀌는 비동기 채널에서 안테나 배열을 쓸 때 유사 최대우도 검파기를 생각했다. 제안한 유사 최대우도 검파기는 먼저 빔을 만든 뒤 상관을 없앤다. 채널 매개변수를 추정하도록 역 여파 신호의 상관행렬 고유분해를 바탕으로 한 방법을 제안하였다. 제안한 알고리듬이 나쁜 전파환경에서도 적은 손실 안에서 채널벡터를 잘 추정하는 것을 모의실험으로 보였다.

• Journal of information and communication convergence engineering
• /
• 제2권4호
• /
• pp.228-232
• /
• 2004

This paper proposes an efficient detection algorithm, which is composed of an interference nulling-and cancelling-based detection algorithm and a maximum likelihood (ML) detection algorithm having reduced numbers of signal points to be tested, for the quasi-orthogonal space-time code with four transmit antennas. When high-level modulation schemes are employed, the algorithm enables the quasi-orthogonal code to achieve near ML performance with a significant reduction in the computational load.

• 음성과학
• /
• 제11권4호
• /
• pp.75-88
• /
• 2004

In this paper, we propose a new approach to sequential linear regression adaptation of continuous density hidden Markov models (CDHMMs) based on transformation space model (TSM). The proposed TSM which characterizes the a priori knowledge of the training speakers associated with maximum likelihood linear regression (MLLR) matrix parameters is effectively described in terms of the latent variable models. The TSM provides various sources of information such as the correlation information, the prior distribution, and the prior knowledge of the regression parameters that are very useful for rapid adaptation. The quasi-Bayes (QB) estimation algorithm is formulated to incrementally update the hyperparameters of the TSM and regression matrices simultaneously. Experimental results showed that the proposed TSM approach is better than that of the conventional quasi-Bayes linear regression (QBLR) algorithm for a small amount of adaptation data.

• Journal of Communications and Networks
• /
• 제13권3호
• /
• pp.257-265
• /
• 2011

In this paper, a cooperative relaying protocol called soft-decision-and-forward (SDF) with multiple antennas in each node is introduced. SDF protocol exploits the soft decision source symbol values from the received signal at the relay node. For orthogonal transmission (OT), orthogonal codes including Alamouti code are used and for non-orthogonal transmission (NT), distributed space-time codes are designed by using a quasi-orthogonal space-time block code. The optimal maximum likelihood (ML) decoders for the proposed protocol with low decoding complexity are proposed. For OT, the ML decoders are derived as symbolwise decoders while for NT, the ML decoders are derived as pairwise decoders. It can be seen through simulations that SDF protocol outperforms AF protocol for both OT and NT.