• Journal of the Korean Data and Information Science Society
• /
• 제9권2호
• /
• pp.165-172
• /
• 1998

This paper considers a linear regression model with censored data where each error term follows a multivariate normal distribution. In this paper we consider the diffuse prior distribution for parameters of the linear regression model. With censored data we derive the full conditional densities for parameters of a multiple regression model in order to obtain the marginal posterior densities of the relevant parameters through the Gibbs Sampler, which was proposed by Geman and Geman(1984) and utilized by Gelfand and Smith(1990) with statistical viewpoint.

• Agribusiness and Information Management
• /
• 제8권2호
• /
• pp.1-8
• /
• 2016

The present study has aimed at analyzing the technical and scale efficiencies of credit utilization by the farmer-borrowers in Chittoor district of Andhra Pradesh, India. DEA approach was followed to analyze the credit utilization efficiency and to analyze the factors influencing the credit utilization efficiency, log-linear regression analysis was attempted. DEA analysis revealed that, the number of farmers operating at CRS are more in number in marginal farms (40%) followed by other (35%) and small (17.5%) farms. Regarding the number of farmers operating at VRS, small farmers dominate the scenario with 72.5 per cent followed by other (67.5%) and marginal (42.5%) farmers. With reference to scale efficiency, marginal farmers are in majority (52.5%) followed by other (47.5%) and small (25%) farmers. At the pooled level, 26.7 per cent of the farmers are being operated at CRS, 63 per cent at VRS and 32.5 per cent of the farmers are either performed at the optimum scale or were close to the optimum scale (farms having scale efficiency values equal to or more than 0.90). Nearly 58, 15 and 28 percents of the farmers in the marginal farms category were found operating in the region of increasing, decreasing and constant returns respectively. Compared to marginal farmers category, there are less number of farmers operating at CRS both in small farmers category (15%) and other farmers category (22.5%). At the pooled level, only 5 per cent of the farmers are operating at DRS, majority of the farmers (73%) are operating at IRS and only 22 per cent of the farmers are operating at CRS indicating efficient utilization of credit. The log-linear regression model fitted to analyze the major determinants of credit utilization (technical) efficiency of farmer-borrowers revealed that, the three variables viz., cost of cultivation and family expenditure (both negatively influencing at 1% significant level) and family income (positively influencing at 1% significant level) are the major determinants of credit utilization efficiency across all the selected farmers categories and at pooled level. The analysis further indicate that, escalation in the cost of cultivation of crop enterprises in the region, rise in family expenditure and prior indebtedness of the farmers are showing adverse influence on the credit utilization efficiency of the farmer-borrowers.

• Journal of the Korean Data and Information Science Society
• /
• 제23권1호
• /
• 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.

• 구강회복응용과학지
• /
• 제32권4호
• /
• pp.280-292
• /
• 2016

목적: 이 연구의 목적은 임플란트의 생존율과 주위 변연골 수준을 후향적으로 평가하여, 술자의 임상적 경험을 포함한 요인들이 미치는 영향을 분석하고자 하였다. 연구 재료 및 방법: 2002년 1월부터 2009년 3월까지 강릉원주대학교치과병원 치주과에서 전공의가 2단계법으로 식립한 임플란트 중 5년 이상의 기록이 있는 146명의 420개 임플란트를 대상으로, 임플란트 탈락여부, 주위 골 소실량, 성별, 연령, 2형 당뇨, 흡연, 지대주 연결 형식, 표면 처리 방법, 직경, 식립 부위, 식립 당시 전공의 연차, 덮개 나사의 조기 노출, 보철학적 합병증, 동반된 골 이식 술식 등의 영향을 평가하였다. 결과: 최종 보철물 장착 후 5년간 누적생존율은 94.9% 이었다. 이원로지스틱회귀분석 결과, 흡연과 덮개 나사의 조기 노출이 실패율을 유의하게 증가시켰다. 다중회귀분석 결과, 흡연, 임플란트 지대주 연결 형식, 및 표면 처리 방법이 임플란트 주위 변연골 수준 변화에 유의한 영향을 주었다. 술자의 임상경험을 반영하기 위한 식립 당시의 전공의 수련 연차는 임플란트의 실패와 유의한 연관성은 보이지 않았다(P = 0.171). 결론: 흡연, 덮개 나사의 조기 노출, 지대주 연결 형식 및 표면 처리 방법이 임플란트의 성공과 유의한 상관관계를 보이는 것으로 생각된다.

• Restorative Dentistry and Endodontics
• /
• 제19권2호
• /
• pp.499-512
• /
• 1994

The purpose of this study was to evaluate the microleakage electrochemically using different retrograde filling materials, different root resection angle, and different cavity preparation instruments. 104 extracted single-rooted teeth were selected for this study. 100 teeth were used as experimental groups and four was used as controls. Anatomical crowns were resected, root canals were prepared, and the apical 2 mm of roots were removed. The experimental roots were randomly divided into five equal groups. Experimental groups : Group 1. no bevel, cavity preparation with ultrasonic instruments, amalgam filling Group 2. no bevel, cavity preparation with ultrasonic instruments, SuperEBA cement filling Group 3. no bevel, cavity preparation with ultrasonic instruments, desiccated ZOE filling Group 4. $45^{\circ}$ bevel, cavity preparation with ultrasonic instruments, amalgam filling Group 5. no bevel, cavity preparation with conventional bur, amalgam filling Microleakage was measured once a day for 30 days using electrochemical method and were analyzed statistically. The results were as follows : 1. The group with Super EBA cement filling showed the least marginal leakage from second to fourth day(p<0.05), there was no significant difference between the group with amalgam filling during eighth to eighteenth day(p>0.05), but after the nineteenth day here was a higher marginal leakage than the group with amalgam filling(p<0.05). 2. The group with desiccated ZOE filling demostrated that the highest marginal leakage, started on the eighth day(p<0.05). 3. The group using ultrasonic instrument showed lower marginal leakage than the group using bur until the nineteenth day(p<0.05), but there was no significant differnce with the group using bur after twentythird day(p>0.05). 4. The group without bevel showed lower marginal leakage than the group with bevel (p<0.05). 5. Whether bevel or nor had much more effect on marginal leakage than with cavity perparation instrument when the cavity was retrogrdefilled with amalgam(stepwise regression).

• Journal of the Korean Statistical Society
• /
• 제19권2호
• /
• pp.145-150
• /
• 1990

In this article we develop the marginal posterior density of the model parameters in the multiresponse regression models when missing values exist only in one response. The resulting density resolves a couple of problems in the estimation approach proposed by Box, Draper, and Hunter (1970) and provides a general interpretation for relationship between the estimates of the missing values and the parameters.

• 한국컴퓨터정보학회논문지
• /
• 제5권3호
• /
• pp.151-158
• /
• 2000

This dissertation measures transitory items using earnings change scaled by beginning-of-period price(FreemanㆍTse 1992) and the earnings-to-price ratio(AliㆍZarowin 1992). Contextual regression model results confirm the incremental explanatory power for predominantly permanent earnings, and suggest that cash flows also have incremental explanatory power in the presence of predominantly permanent earnings. But contextual regression results represent that while earnings are consistent with a smaller marginal impact from extreme (transitory) earnings on abnormal returns, cash flows have no greater impact on abnormal returns in the presence of large transitory components in earnings.

• Journal of the Korean Statistical Society
• /
• 제29권4호
• /
• pp.407-422
• /
• 2000

This article is concerned with the selecting predictor variables to be included in building a class of binary response t-link regression models where both probit and logistic regression models can e approximately taken as members of the class. It is based on a modification of the stochastic search variable selection method(SSVS), intended to propose and develop a Bayesian procedure that used probabilistic considerations for selecting promising subsets of predictor variables. The procedure reformulates the binary response t-link regression setup in a hierarchical truncated normal mixture model by introducing a set of hyperparameters that will be used to identify subset choices. In this setup, the most promising subset of predictors can be identified as that with highest posterior probability in the marginal posterior distribution of the hyperparameters. To highlight the merit of the procedure, an illustrative numerical example is given.

• 한국해양학회지:바다
• /
• 제23권2호
• /
• pp.63-75
• /
• 2018

본 연구에서는 분위회귀를 이용하여 1993~2016년 동안의 한반도 조위 자료의 장주기 선형 추세를 분석하였다. 일반 선형회귀(OLS: Ordinary Least Square)와 50% 중간 분위 회귀 결과를 비교했을 때 특히 황해에서 약 2~3 mm/year의 회귀 결과의 차이를 발견하였으며, 이는 극한 값 변화에 기인함을 확인할 수 있었다. 또한 해역별로 하위(1%) 분위와 중간(50%) 분위, 상위(99%) 분위의 값이 모두 다르게 나타났다. 황해의 대부분 지역에서는 상위 분위와 하위 분위가 모두 증가하는 경향을 나타냈으나, 유의미한 "상향 발산형" 회귀 결과를 보였다. 이는 상위 분위가 중간 분위에 비해 유의미하게 크게 나타나는 경향을 의미한다. 대한민국 남해안에서는 상위 분위가 하위 분위보다 더 작은 증가 값을 가지는 "상향 수렴형" 회귀 결과가 특징적으로 나타났다. 이러한 경향은 조화 분석을 통해 알려진 조석 분조들을 제거한 결과에서는 없어지는 것을 확인하였다. 그러므로 분위 회귀의 지역적 특성은 조석 세기의 장주기 변동과 연관이 있다고 추측된다.

• 응용통계연구
• /
• 제28권6호
• /
• pp.1191-1208
• /
• 2015

대부분의 통계분석방법은 요약통계량에 의존하지만 그래픽적 방법을 이용하면 자료의 특성을 파악하기 쉽고 통계량만으로는 알아낼 수 없는 부분까지도 접근이 가능하다. 그래프를 통한 로지스틱회귀모형의 평가 방법으로 로그-밀도비를 통한 검토, 차원 검토, 주변모형산점도, 카이잔차산점도, CERES 그림을 알아보고 모의자료들을 통해 다양한 상황에서 그래픽적 방법들 어떠한 결과를 나타내지를 비교 검토한다.