• Communications for Statistical Applications and Methods
• /
• v.9 no.3
• /
• pp.809-824
• /
• 2002

we propose two methods which separate the variable selection step and the split-point selection step. We call these two algorithms as CHITES method and F&CHITES method. They adapted some of the best characteristics of CART, CHAID, and QUEST. In the first step the variable, which is most significant to predict the target class values, is selected. In the second step, the exhaustive search method is applied to find the splitting point based on the selected variable in the first step. We compared the proposed methods, CART, and QUEST in terms of variable selection bias and power, error rates, and training times. The proposed methods are not only unbiased in the null case, but also powerful for selecting correct variables in non-null cases.

• The Korean Journal of Applied Statistics
• /
• v.14 no.2
• /
• pp.475-486
• /
• 2001

데이터마이닝 패키지에 구현된 분류나무 알고리즘 가운데 CART, CHAID, QUEST, C4.5에서 변수 선택법을 비교하였다. CART의 전체탐색법이 편의를 갖는다는 사실은 잘알려졌으며, 여기서는 상품화된 패키지들에서 이들 알고리즘의 편의와 선택력을 모의실험 연구를 통하여 비교하였다. 상용 패키지로는 CART, Enterprise Miner, AnswerTree, Clementine을 사용하였다. 본 논문의 제한된 모의실험 연구 결과에 의하면 C4.5와 CART는 모두 변수선택에서 심각한 편의를 갖고 있으며, CHAID와 QUEST는 비교적 안정된 결과를 보여주고 있었다.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.5
• /
• pp.1099-1107
• /
• 2017

In this study, we analyzed the determinants of wages of college graduates by using the data of "2014 Graduates Occupational Mobility Survey" conducted by Korea Employment Information Service. In general, wages contain two complex pieces of information about whether an individual is employed and the size of the wage. However, in many previous researches on wage determinants, sample selection bias tends to be generated by performing linear regression analysis using only information on wage size. We used the Heckman sample selection models for analysis to overcome this problem. The main results are summarized as follows. First, the validity of the Heckman's sample selection model is statistically significant. Male is significantly higher in both job probability and wage than female. As age increases and parents' income increases, both the probability of employment and the size of wages are higher. Finally, as the university satisfaction increases and the number of certifications acquired increased, both the probability of employment and the wage tends to increase.

• The Korean Journal of Applied Statistics
• /
• v.29 no.5
• /
• pp.849-858
• /
• 2016

It is important to control confounding bias when estimating the causal effect of treatment in an observational study. We illustrated that the covariate selection in the causal inference is different from the variable selection in the ANCOVA model. We then investigated the three criteria of covariate selection for controlling confounding bias, which can be used when we have inadequate information to draw a complete causal graph. VanderWeele and Shpitser (2011) proposed one of them and claimed it was better than the other two. We show by example that their criterion also has limitations and some disadvantages. There is no clear winner; however, their criterion is better (if some correction is made on its condition) than the other two because it can remove the confounding bias.

• Proceedings of the Korean Statistical Society Conference
• /
• 2001.11a
• /
• pp.89-94
• /
• 2001

In this study we focus on variable selection in decision tree growing structure. Some of the splitting rules and variable selection algorithms are discussed. We propose a competitive variable selection method based on Kruskal-Wallis test, which is a nonparametric version of ANOVA F-test. Through a Monte Carlo study we note that CART has serious bias in variable selection towards categorical variables having many values, and also QUEST using F-test is not so powerful to select informative variables under heavy tailed distributions.

• English Language & Literature Teaching
• /
• v.16 no.1
• /
• pp.123-142
• /
• 2009

The present study used multi-faceted Rasch measurement to explore the characteristics and bias patterns of non-native raters when they scored L2 writing tasks. Three raters scored 254 writing tasks written by Korean university students on two topics adapted from the TOEFL Test of Written English (TWE). The written products were assessed using a five-category rating scale (Content, Organization, Language in Use, Grammar, and Mechanics). The raters only showed a difference in severity with regard to rating categories but not in task types. Overall, the raters scored Grammar most harshly and Organization most leniently. The results also indicated several bias patterns of ratings with regard to the rating categories and task types. In rater-task bias interactions, each rater showed recurring bias patterns in their rating between two writing tasks. Analysis of rater-category bias interaction showed that the three raters revealed biased patterns across all the rating categories though they were relatively consistent in their rating. The study has implications for the importance of rater training and task selection in L2 writing assessment.

• Survey Research
• /
• v.7 no.2
• /
• pp.39-52
• /
• 2006

Most of Korean survey institutions adopt quota sampling for telephone surveys based on region, gender and age-band. In weekdays, it is well blown that there exist substantial differences in day time in-house rate by individual's socio-demographic attributes. So, quota sampling may induce systematic respondent selection bias. To solve the problem, we propose "time-balanced quota sampling" in which interviewer's call time-band is added as an quota variable. Furthermore, we propose "time-balanced quasi-quota sampling" which is derived by partially relaxing evening time quotas in time-balanced quota sampling. We compare the conventional and the newly proposed quota sampling schemes by drawing Monte Carlo samples from the hypothetical population for which the Korea 2004 time use survey data is assumed.

• The Journal of Asian Finance, Economics and Business
• /
• v.9 no.2
• /
• pp.347-353
• /
• 2022

Using the Heckman two-step procedure, this study investigates the relationship between management accounting implementation and firm effectiveness. The research data for this study was acquired from 450 publicly traded companies in Vietnam; however, the final sample only includes 304 responses containing useful information. The reliability analysis was used to evaluate the acquired data to examine the qualities of constructs and the dimensions that make them up. Then, the Heckman two-step technique was performed to analyze the causal connection from the acceptance of management accounting to firm effectiveness allowing for the effect of environmental uncertainty and organizational characteristics on the likelihood of adopting management accounting. The empirical findings show that management accounting acceptance determines firm effectiveness; however, the research model on the relationship between management accounting adoption and firm effectiveness has a sample selection bias. The main conclusions of this study are that there is a difference in the effects of management accounting adoption on business effectiveness when sample selection bias is not taken into consideration. When potential sample selection bias is taken into account by integrating environmental uncertainty and organizational characteristics in the research model, the effect of adopting management accounting on company effectiveness becomes minor.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.10
• /
• pp.73-81
• /
• 2015

The paper proposes a prototype selection method and evaluates the generalization performance of standard algorithms and prototype based classification learning. The proposed prototype classifier defines multidimensional spheres with variable radii within class areas and generates a small set of training data. The nearest-neighbor classifier uses the new training set for predicting the class of test data. By decomposing bias and variance of the mean expected error value, we compare the generalization errors of k-nearest neighbor, Bayesian classifier, prototype selection using fixed radius and the proposed prototype selection method. In experiments, the bias-variance changing trends of the proposed prototype classifier are similar to those of nearest neighbor classifiers with all training data and the prototype selection rates are under 27.0% on average.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.1
• /
• pp.112-127
• /
• 2002

A useful pattern is a pattern that contributes much to learning. For a classification problem those patterns near the class boundary surfaces carry more information to the classifier. For a regression problem the ones near the estimated surface carry more information. In both cases, the usefulness is defined only for those patterns either without error or with negligible error. Using only the useful patterns gives several benefits. First, computational complexity in memory and time for learning is decreased. Second, overfitting is avoided even when the learner is over-sized. Third, learning results in more stable learners. In this paper, we propose a pattern 'utility index' that measures the utility of an individual pattern. The utility index is based on the bias and variance of a pattern trained by a network ensemble. In classification, the pattern with a low bias and a high variance gets a high score. In regression, on the other hand, the one with a low bias and a low variance gets a high score. Based on the distribution of the utility index, the original training set is divided into a high-score group and a low-score group. Only the high-score group is then used for training. The proposed method is tested on synthetic and real-world benchmark datasets. The proposed approach gives a better or at least similar performance.