• Journal of the Korean Data and Information Science Society
• /
• v.25 no.1
• /
• pp.177-185
• /
• 2014

Under bivariate normal distribution assumptions, the interaction and quadratic terms are needed in the logistic regression model with two predictors. However, depending on the correlation coefficient and the variances of two conditional distributions, the interaction and quadratic terms may not be necessary. Although the need for these terms can be determined by comparing the two scatter plots, it is not as useful for interaction terms. We explore the structure and usefulness of the 3-D residual plot as a tool for dealing with interaction in logistic regression models. If predictors have an interaction effect, a 3-D residual plot can show the effect. This is illustrated by simulated and real data.

• The Korean Journal of Applied Statistics
• /
• v.26 no.1
• /
• pp.141-149
• /
• 2013

We present methods for studying the log-density ratio that enables the selection of the predictors and the form to be included in the logistic regression model. Under bivariate normal distributional assumptions, we investigate the form of the log-density ratio as a function of two predictors. If two covariance matrices are equal, then the crossproduct and quadratic terms are not needed. If the variables are uncorrelated, we do not need the crossproduct terms, but we still need the linear and quadratic terms. We also explore other conditions in which the crossproduct and quadratic terms are not needed in the logistic regression model.

• The Korean Journal of Applied Statistics
• /
• v.27 no.2
• /
• pp.277-290
• /
• 2014

We analyzed binary landslide data from the Boeun area with logistic regression. Since the number of landslide occurrences is only 9 out of 5000 observations, this can be regarded as a rare event data. The main issue of logistic regression with the rare event data is a serious bias problem in regression coefficient estimates. Two bias correction methods were proposed before and we quantitatively compared them via simulation. Firth (1993)'s approach outperformed and provided the most stable results for analyzing the rare-event binary data.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.1
• /
• pp.125-133
• /
• 2013

In this paper, penalized binary logistic regression models are employed as statistical models for determining the discharge of 668 patients with a chief complaint of dyspnea based on 11 blood tests results. Specifically, the ridge model based on $L^2$ penalty and the Lasso model based on $L^1$ penalty are considered in this paper. In the comparison of prediction accuracy, our models are compared with the logistic regression models with all 11 explanatory variables and the selected variables by variable selection method. The results show that the prediction accuracy of the ridge logistic regression model is the best among 4 models based on 10-fold cross-validation.

• The Korean Journal of Applied Statistics
• /
• v.14 no.1
• /
• pp.71-80
• /
• 2001

Logistic regression model is one of the most popular linear models for a binary response variable and used for the estimation of probability function. In many practical situations, the probability function can be expressed by a bell shaped curve and such a function can be estimated by a second order logistic regression model. However, when the probability curve is asymmetric, the estimation results using a second order logistic regression model may not be precise because a second order logistic regression model is a symmetric function. In addition, even if a second order logistic regression model is used, the interpretation for the effect of second order term may not be easy. In this paper, in order to alleviate such problems, an estimation method for asymmetric probabiity curve based on a first order logistic regression model and iterative bi-section method is proposed and its performance is compared with that of a second order logistic regression model by a simulation study.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.354-354
• /
• 2021

2020년은 장마기간이 49일간 지속됨에 따라 침수, 산사태 등 많은 홍수피해가 발생하였다. 특히 서울에서는 한강 본류의 수위가 급격하게 증가함에 따라 둔치 및 도로 침수 피해가 발생하였다. 이처럼 하천의 수위증가로 인한 홍수피해에 대응하기 위해 홍수통제소 및 기초지자체에서는 홍수특보를 발령한다. 이 홍수특보는 수위관측소 지점별 계획홍수량의 50 %, 70 % 이상의 홍수량이 발생할 경우 홍수주의보와 홍수경보가 발령되며, 이 기준은 각 권역별로 동일하다. 하지만 2017년 의정부시에서는 중랑천 수위증가로 인해 주변 지역에 침수피해가 발생하였지만, 이때 홍수량은 계획홍수량 대비 약 30 %에 불과하였다. 이처럼 한강권역 내 하천수위 증가로 인한 홍수피해는 계획홍수량의 50 % 이내에서 발생하기도 한다. 이에 본 연구에서는 한강권역을 대상으로 현재 2단계로 발령되는 홍수특보를 3단계로 세분화하고자 하였다. 단계별 홍수량 위험기준을 산정하기 위해 과거 홍수피해 발생 이력이 있는 한강권역 내 43개의 수위관측소 지점을 선정하였으며, 지점별 홍수기 동안의 홍수량 및 피해액 자료를 수집하였다. 각 단계별 홍수량 기준을 산정하기 위해서는 로지스틱 회귀분석 방법을 활용하여 피해발생 확률을 산정하였다. 1단계 기준은 계획홍수량 대비 홍수량 비율과 홍수피해 발생여부를 고려한 이항 로지스틱 회귀분석 모델을 구축한 후 3계 도함수에 적용하여 홍수피해 발생확률이 급격하게 증가하는 특이점을 산정하였다. 2단계와 3단계 기준은 다항 로지스틱 회귀분석 중 계층형 로지스틱 회귀분석을 활용하여 지점별 피해액 비율이 60 ~ 80 %, 80 ~ 100 % 구간에 속할 확률을 산정하고, 1단계와 동일한 방법으로 특이점을 산정하였다. 그 결과 지점별로 기존 제공되고 있는 홍수특보 기준을 과거 발생한 홍수피해를 고려하여 세분화할 수 있었으며, 이 결과는 지역별 홍수피해 저감대책에 활용될 수 있을 것으로 판단된다.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.1
• /
• pp.21-32
• /
• 2010

Graphical regression is a paradigm for obtaining regression information using plots without model assumptions. The general goal of this approach is to find lowdimensional sufficient summary plots without loss of important information. Model assessments using residual plots are less likely to be successful in models that are not linear. As an alternative approach, marginal model plots provide a general graphical method for assessing the model. We apply the methods of graphical regression and model assessment using marginal model plots to the logistic regression model.

• Communications for Statistical Applications and Methods
• /
• v.19 no.1
• /
• pp.1-11
• /
• 2012

We present methods to study the log-density ratio of the conditional densities of the predictors given the response variable in the logistic regression model. This allows us to select which predictors are needed and how they should be included in the model. If the conditional distributions are skewed, the distributions can be considered as gamma distributions. A simulation study shows that the linear and log terms are required in general. If the conditional distributions of xjy for the two groups overlap significantly, we need both the linear and log terms; however, only the linear or log term is needed in the model if they are well separated.

• Journal of the Korea Society for Simulation
• /
• v.27 no.1
• /
• pp.51-60
• /
• 2018

This study aims to understand crucial factors affecting user's Fintech payment service adoption. On the basis of innovation diffusion theory and prior Fintech literature, this study classifies the influence factors of users' adoption of Fintech payment service into two dimensions - service dimension containing complexity, perceived benefit, trust in service provider and user dimension containing personal innovativeness and security breach experience. The data analysis results using binary logistic regression shows the negative direct effects of perceived risk, complexity, security accident experience on user's service adoption are statistically significant. Personal innovativeness has a positive effect on user's Fintech payment service adoption. The moderation effect of security accident experience is also significant at p<0.05.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.681-690
• /
• 2017

Various imbalanced binary classification problems exist such as fraud detection in banking operations, detecting spam mail and predicting defective products. Several sampling methods such as over sampling, under sampling, SMOTE have been developed to overcome the poor prediction performance of binary classifiers when the proportion of one group is dominant. In order to overcome this problem, several sampling methods such as over-sampling, under-sampling, SMOTE have been developed. In this study, we investigate prediction performance of logistic regression, Lasso, random forest, boosting and support vector machine in combination with the sampling methods for binary imbalanced data. Four real data sets are analyzed to see if there is a substantial improvement in prediction performance. We also emphasize some precautions when the sampling methods are implemented.