• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.277-289
• /
• 2003

We propose modified exact inferential methods in logistic regression model. Exact conditional distribution in logistic regression model is often highly discrete, and ordinary exact inference in logistic regression is conservative, because of the discreteness of the distribution. For the exact inference in logistic regression model we utilize the modified P-value. The modified P-value can not exceed the ordinary P-value, so the test of size $\alpha$ based on the modified P-value is less conservative. The modified exact confidence interval maintains at least a fixed confidence level but tends to be much narrower. The approach inverts results of a test with a modified P-value utilizing the test statistic and table probabilities in logistic regression model.

• Communications for Statistical Applications and Methods
• /
• 제24권4호
• /
• pp.339-351
• /
• 2017

Resampling approaches were the first techniques employed to compute a variance for the Gini coefficient; however, many authors have shown that an analysis of the Gini coefficient and its corresponding variance can be obtained from a regression model. Despite the simplicity of the regression approach method to compute a standard error for the Gini coefficient, the use of the proposed regression model has been challenging in economics. Therefore in this paper, we focus on a comparative study among the regression approach and resampling techniques. The regression method is shown to overestimate the standard error of the Gini index. The simulations show that the Gini estimator based on the modified regression model is also consistent and asymptotically normal with less divergence from normal distribution than other resampling techniques.

• 한국산업정보학회논문지
• /
• 제4권2호
• /
• pp.39-43
• /
• 1999

임의중도절단자료(randomly censored data)를 갖는 일반화된 지수회귀모헝을 고려하여 이 모형에서 모수를 추정하는 수정된 피선 점수화(modified Fisher scoring)방법을 제안한다. 이를 위해 우도방정식(likelihood equations)이 유도되고 추정알고리즘(estimating algorithm)이 개발된다. 실제의 자료를 통해 제안된 방법을 예증한다.

• 경영과정보연구
• /
• 제7권
• /
• pp.427-450
• /
• 2001

This research suggests the estimation methodology of Logistics. This paper elucidates the main problems associated with estimation in the regression model. We review the methods for estimating the parameters in the model and introduce a modified procedure in which all models are fitted and combined to construct a combination of estimates. The resulting estimators are found to be as efficient as the maximum likelihood (ML) estimators in various cases. Our method requires more computations but has an advantage for large data sets. Also, it enables to detect particular features in the data structure. Examples of real data are used to illustrate the properties of the estimators. The backgrounds of estimation of logistic regression model is the increasing logistic environment importance today. In the first phase, we conduct an exploratory study to discuss 9 independent variables. In the second phase, we try to find the fittest logistic regression model. In the third phase, we calculate the logistic estimation using logistic regression model. The parameters of logistic regression model were estimated using ordinary least squares regression. The standard assumptions of OLS estimation were tested. The calculated value of the F-statistics for the logistic regression model is significant at the 5% level. The logistic regression model also explains a significant amount of variance in the dependent variable. The parameter estimates of the logistic regression model with t-statistics in parentheses are presented in Table. The object of this paper is to find the best logistic regression model to estimate the comparative accurate logistics.

• 한국식생활문화학회지
• /
• 제23권4호
• /
• pp.421-426
• /
• 2008

The purpose of this study was to measure the effects of trust, knowledge, optimism, risk and benefits on consumer attitudes toward genetically modified foods. A total of 326 questionnaires were completed. Moderated regression analysis was used to measure the relationships among the variables. The analysis results for the data indicated a good model fit in Model 2 rather than Model 1, in which the direct effects of trust, optimism and benefits had statistically significant direct effects on the respondents' attitudes toward genetically modified foods, while the direct effects of knowledge and risk were not statistically significant. As expected, the interaction term of risk and benefit had a significant effect on consumer attitude. Moreover, the effect of risk on consumer's attitude toward genetically modified foods was statistically significant at all levels of benefit, except at the lower benefit level. Finally, the results of this study indicated that genetically modified food developers and marketers should attach importance to the interaction effect of benefits to understand the elements of market demand and customer loyalty.

• 자원환경지질
• /
• 제40권1호
• /
• pp.115-128
• /
• 2007

본 연구는 화강암질암 지역의 자연사면에서 발생하는 토석류 산사태의 발생지점을 확률론적 예측하기 위하여 기 개발된 로지스틱 회귀모델을 수정하고자 한다. 기 모델의 단점인 일부 범주형 변수사용을 제거하여 예측률의 신뢰도 및 예측도면 작성시의 정확성을 높인 새로운 예측모델을 제안하고자 한다. 새롭게 개발된 모델은 암상, 지형인자 2개 및 토질인자 3개를 사용하여 통계적으로 86%이상의 예측률을 확보하였다. 본 모델의 적용성을 검증하기 위하여 태풍 '루사'로 인해 산사태가 집중적으로 발생한 강릉지역에 적용하여 산사태 예측도를 작성하였다. 예측결과 사천지역의 경우 본 모델에서 고려하지 못한 산불의 영향으로 산불피해지역에서 근소한 차이를 보여주고 있으나, 주문진-연곡지역의 경우는 예측결과가 실제 산사태 발생위치와 잘 일치하고 있다. 따라서 본 모델은 우리나라의 화강암질암지역에 적용하여 널리 활용될 수 있을것으로 판단된다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.431-434
• /
• 2003

In this paper we introduce a modified objective function for fuzzy c-means clustering with logistic regression model in the presence of noise cluster. The logistic regression model is commonly used to describe the effect of one or several explanatory variables on a binary response variable. In real application there is very often no sharp boundary between clusters so that fuzzy clustering is often better suited for the data.

• The Journal of Asian Finance, Economics and Business
• /
• 제7권8호
• /
• pp.533-542
• /
• 2020

This study estimates the nature of the relationship of entrepreneurship and business confidence on youth unemployment in the Philippines over the 2001-2017 period. The paper employed a range of cointegrating regression models, namely, autoregressive distributed lag (ARDL) bounds testing approach, Johansen-Juselius (JJ) and Engle-Granger (EG) cointegration models, dynamic OLS, fully modified OLS, and canonical cointegrating regression (CCR) estimation techniques. The Granger causality based on error correction model (ECM) was also performed to determine the causal link of entrepreneurship and business confidence on youth unemployment. The ARDL bounds testing approach, Johansen-Juselius (JJ) and Engle-Granger (EG) cointegration models confirmed the existence of long-run equilibrium relationship of entrepreneurship and business confidence on youth unemployment. The long-run coefficients from JJ and dynamic OLS show significant long-run and positive relationship of entrepreneurship and business confidence on youth unemployment. While results of the long-run coefficients from fully modified OLS and canonical cointegrating regression (CCR) found that only entrepreneurship has significant and positive relationship with youth unemployment in the long-run. The Granger causality based on error correction model (ECM) estimates show evidence of long-run causal relationship of entrepreneurship and business confidence on youth unemployment. In the short-run, increases in entrepreneurship and business confidence causes youth unemployment to decrease.

• 응용통계연구
• /
• 제22권1호
• /
• pp.197-204
• /
• 2009

In the model selection problem, the main objective is to choose the true model from a manageable set of candidate models. An information criterion gauges the validity of a statistical model and judges the balance between goodness-of-fit and parsimony; "how well observed values ran approximate to the true values" and "how much information can be explained by the lower dimensional model" In this study, we introduce some information criteria modified from the Akaike Information Criterion (AIC) and the Bayesian Information Criterion(BIC). The information criteria considered in this study are compared via simulation studies and real application.

• 대한토목학회논문집
• /
• 제34권4호
• /
• pp.1139-1150
• /
• 2014

본 연구에서는 M-PRISM 모형을 이용하여 $1km{\times}1km$ 공간해상도 일강수량 추정에 대한 적용성을 검토하였다. 또한 회귀모형을 이용하여 M-PRISM 모형 매개변수를 추정하였으며, 잭나이프 방법을 이용하여 모형을 검증하였다. 기상청 385개 강수 관측지점에 대하여 M-PRISM을 이용하여 일강수량을 추정하고 PRISM 모형과 비교하였다. 비교결과, 강수의 정량적 크기를 추정에서는 두 모형에서 뚜렷한 차이를 찾아볼 수 없었으나, 강수의 발생빈도 추정에 있어서는 M-PRISM 모형이 더 우수한 결과를 나타내었다. 따라서 본 연구에서 제안한 M-PRISM 모형은 고해상도의 일강수량을 추정함에 있어서 매우 유용하게 사용될 수 있을 것으로 판단된다.