• Communications for Statistical Applications and Methods
• /
• v.10 no.3
• /
• pp.971-980
• /
• 2003

The tree method can be extended to multivariate responses, such as repeated measure and longitudinal data, by modifying the split function so as to accommodate multiple responses. Recently, some decision trees for multiple responses have been constructed by Segal (1992) and Zhang (1998). Segal suggested a tree can analyze continuous longitudinal response using Mahalanobis distance for within node homogeneity measures and Zhang suggested a tree can analyze multiple binary responses using generalized entropy criterion which is proportional to maximum likelihood of joint distribution of multiple binary responses. In this paper, we will modify CART procedure and suggest a new tree-based method that can analyze multiple binary responses using similarity measures.

• Communications for Statistical Applications and Methods
• /
• v.9 no.3
• /
• pp.703-713
• /
• 2002

In longitudinal studies, outcomes are repeatedly measured over time for each subject. It is common to have missing values or dropouts for longitudinal data. In this study time trend in longitudinal data with dropouts is of concern. The confounding of time trend with dropout process is investigated through simulation studies. Some simulation results are reported for binary responses as well as continuous responses with patterns of dropouts varying. It has been found that time trend is not confounded with random dropout process for binary responses when it is estimated using GEE.

• The Korean Journal of Applied Statistics
• /
• v.14 no.2
• /
• pp.321-332
• /
• 2001

본 논문에서는 반응변수가 두 가지의 값을 갖는 회귀분석에 적용할 수 있는 사영추적회귀를 고려하였다. 회귀모형에 필요한 설명변수들의 선형결합이 하나이고 연결함수의 형태를 사전에 알지 못한다는 가정하에서 모의담금질 기법을 이용하여 모형에 필요한 선형결합을 찾는 알고리즘을 제시하였다. 이진 반응변수의 경우에는 평활모수의 값에 따라 잔차이탈도함수의 반응표면이 단봉의 형태를 갖지 않는 경우가 있어 비동질적 마코프체인을 이용한 모의담금질 기법을 적용하면 효율적으로 선형결합을 탐색할 수 있다.

• Korean Journal of Veterinary Research
• /
• v.25 no.2
• /
• pp.155-165
• /
• 1985

Effects of binary ethylenimine (BEI) treatment on the inactivation of infectivity and hemagglutinin of Newcastle disease virus (NDV) were studied in comparison with those of formalin treatment. Immune responses of chickens vaccinated with BEI-inactivated NDV vaccines were also investigated. The results were summarized as followings; 1. Complete loss of infectivity of NDV (Bl) was observed at 3, 7, and 24 hours after the treatment at $37^{\circ}C$ with BEI concentrations of 0.01M, 0.005M and 0.001M, respectively. 2. The hemagglutinin activity of NDV (Bl) remained constant when treated with 0.01M BEI at $37^{\circ}C$. However, it gradually decreased when treated with 0.1% or 0.2% formalin at $37^{\circ}C$. 3. When 4-week-old chickens were vaccinated with NDV vaccines prepared from Bl or Miyadera strains of NDV, inactivated with 0.1M BEI and adsorbed to aluminium hydroxide gel, favorable immune responses were observed throughout the 8 weeks of observation period. 4. When these chickens were revaccinated at 8 weeks after the first vaccination, strong anamnestic responses were evoked and the immunity maintained for 4 weeks of the observation. Though slightly bettor immune responses were observed after primary vaccination in chickens vaccinated with Bl vaccine compared with those vaccinated with Miyadera vaccine, the differences were not significant. 5. On the electron microscopy, BEI (0.01M) gave least effect to the envelope as well as capsid of NDV.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.1
• /
• pp.209-218
• /
• 2012

The method of generalized estimating equations (GEEs) provides consistent esti- mates of the regression parameters in a marginal regression model for longitudinal data, even when the working correlation model is misspecified (Liang and Zeger, 1986). In this paper we compare the estimators of parameters in GEE approach. We consider two aspects: coverage probabilites and efficiency. We adopted to ordinal responses th results derived from binary outcomes.

• Journal of the Korean Statistical Society
• /
• v.30 no.4
• /
• pp.613-635
• /
• 2001

This paper proposes a skewed multivariate probit model for analyzing a correlated binary response data with covariates. The proposed model is formulated by introducing an asymmetric link based upon a skewed multivariate normal distribution. The model connected to the asymmetric multivariate link, allows for flexible modeling of the correlation structure among binary responses and straightforward interpretation of the parameters. However, complex likelihood function of the model prevents us from fitting and analyzing the model analytically. Simulation-based Bayesian inference methodologies are provided to overcome the problem. We examine the suggested methods through two data sets in order to demonstrate their performances.

• Journal of the Korean Statistical Society
• /
• v.26 no.4
• /
• pp.493-505
• /
• 1997

The analysis of binary data appears to many areas such as statistics, biometrics and econometrics. In many cases, data are often collected in which some observations are incomplete. Assume that the missing covariates are missing at random and the responses are completely observed. A method to Bayesian analysis of the binary regression model with incomplete data is presented. In particular, the desired marginal posterior moments of regression parameter are obtained using Meterpolis algorithm (Metropolis et al. 1953) within Gibbs sampler (Gelfand and Smith, 1990). Also, we compare logit model with probit model using Bayes factor which is approximated by importance sampling method. One example is presented.

• The Korean Journal of Applied Statistics
• /
• v.37 no.4
• /
• pp.467-484
• /
• 2024

Longitudinal data are measured repeatedly over time from the same subject, so there is a correlation from the repeated outcomes. Therefore, when analyzing this correlation, both serial correlation and between-subject variation must be considered in longitudinal data analysis. In this paper, we will focus on the marginalized models to estimate the population average effect of covariates among models for analyzing longitudinal binary data. Marginalized models for longitudinal binary data include marginalized random effects models, marginalized transition models, and marginalized transition random effect models, and in this paper, these models are first reviewed, and simulations are conducted using complete data and missing data to compare the performance of the models. When there were missing values in the data, there is a difference in performance depending on the model in which the data was generated. We analyze Korea Health Panel data using marginalized models. The Korean Medical Panel data considers subjective unhealthy responses as response variables as binary variables, compares models with several explanatory variables, and presents the most suitable model.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.1
• /
• pp.187-194
• /
• 2014

We consider a Bayesian nonignorable selection model to accommodate the selection bias. Markov chain Monte Carlo methods is known to be very useful to fit the nonignorable selection model. However, sensitivity to prior assumptions on parameters for selection mechanism is a potential problem. To quantify the sensitivity to prior assumption, the deviance information criterion and the conditional predictive ordinate are used to compare the goodness-of-fit under two different prior specifications. It turns out that the 'MLE' prior gives better fit than the 'uniform' prior in viewpoints of goodness-of-fit measures.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 1999.03a
• /
• pp.281-284
• /
• 1999

본 연구는 인간의 뇌파로부터 간단한 의사 표시를 식별하는 기술을 얻어 뇌파인터페이스를 구현하기 위한 기초연구로서 수행되었다. 실험에 참가한 피험자들은 컴퓨터 화면에 나타나는 문제를 본 후 답을 제시받았을 때 이것이 옳은지, 그른지에 대한 양분법적 판단반응을 해야하며, 이때 동시에 뇌파가 기록되었다. 옳다는 긍정반응과, 옳지 않다는 부정반응시의 뇌파를 비교한 결과 전두엽 부위의 fp1, f3, f4 부위에서 부정의 대답을 할 경우 theta파와 fast alpha파의 상대적 출현량이 긍정의 경우에 비하여 통계적으로 유의하게 컸다.