• Communications for Statistical Applications and Methods
• /
• 제9권2호
• /
• pp.521-531
• /
• 2002

Various diagnostic techniques for identifying influential observations are mostly based on the deletion of a single observation. While such techniques can satisfactorily identify influential observations in many cases, they will not always be successful because of some mask effect. It is necessary, therefore, to develop techniques that examine the potentially influential effects of a subset of observations. The partial regression plots can be used to examine an influential observation for a single parameter in multiple linear regression. However, it is often desirable to detect influential observations for a subset of regression parameters when interest centers on a selected subset of independent variables. Thus, we propose a diagnostic measure which deals with detecting influential observations on a subset of regression parameters. In this paper, we propose a measure M, which can be effectively used for the detection of influential observations on a subset of regression parameters in multiple linear regression. An illustrated example is given to show how we can use the new measure M to identify influential observations on a subset of regression parameters.

• Communications for Statistical Applications and Methods
• /
• 제9권1호
• /
• pp.197-212
• /
• 2002

When assessing goodness-of-fit of a model, a small subset of deviating observations can give rise to a significant lack of fit. It is therefore important to identify such observations and to assess their effects on various aspects of analysis. A Cook's distance measure is usually used to detect influential observation. But it sometimes is not fully effective in identifying truly influential set of observations because there may exist masking or swamping effects. In this paper we confine our attention to influential subset In GLMs such as logistic regression models and loglinear models. We modify a backwards stepping algorithm, which was originally suggested for detecting outlying cells in contingency tables, to detect influential observations in GLMs. The algorithm consists of two steps, the identification step and the testing step. In identification step we Identify influential observations based on influencial measures such as Cook's distances. On the other hand in testing step we test the subset of identified observations to be significant or not Finally we explain the proposed method through two types of dataset related to logistic regression model and loglinear model, respectively.

• 품질경영학회지
• /
• 제16권2호
• /
• pp.18-33
• /
• 1988

A new diagnostic statistic for detecting outliers and influential observations in linear models is suggested and studied in this paper. The proposed statistic is a weighted sum of two measures ; one is for detecting outliers and the other is for detecting influential ovservations. The merit of this statistic is that it is possible to distinguish outliers from influential observations. This statistic can be used for not only regression models but also factorial design models. A Monte Carlo simulation study is reported to suggest critical values for detecting outliers and influential observations for simple regression models when the number of observations is 11. 21, 31, 41 or 51.

• Journal of applied mathematics & informatics
• /
• 제5권2호
• /
• pp.489-498
• /
• 1998

The influence of observations the on the goodness-of-fit test in maximum likelihood factor analysis is investigated by using the local influence method. under an appropriate perturbation the test statistic forms a surface. One of main diagnostics is the maximum slope of the perturbed surface the other is the direction vector cor-responding to the curvature. These influence measures provide the information about jointly influence measures provide the information about jointly influential observations as well as individ-ually influential observations.

• Journal of the Korean Data and Information Science Society
• /
• 제28권6호
• /
• pp.1291-1300
• /
• 2017

고차원 자료에서는 관측값의 개수보다 변수의 개수가 과다하게 많은 것이 특징이다. 그러므로 회귀 계수 추정에 있어 관측값의 영향이 매우 클 수 있다. Jang과 Anserson-Cook (2017)은 라쏘추정량 사용시 영향점의 영향을 평가할 수 있는 라쏘 영향그림을 제안하였다. 본 연구에서는 고차원 자료에서 영향점을 평가하기 위한 그래픽 방법들로서 라쏘 영향그림 뿐만 아니라 라쏘 변수선택 순위그림, 삼차원 라쏘 영향그림을 제안하였다. 실세 두 가지 고차원 자료 예들에 영향점들을 찾기 위한 회귀진단 수단으로서 세가지 그래픽 방법들을 사용하여 본 결과 영향점들을 효과적으로 찾아낼 수 있었다.

• Communications for Statistical Applications and Methods
• /
• 제8권2호
• /
• pp.347-355
• /
• 2001

We propose a method for detecting influential observations that have a large influence on the likelihood ratio test statistic for the two sets of variables are uncorrelated with one another. For this purpose we derive a local influence measure for the likelihood ratio test statistic under certain perturbation scheme. An illustrative example is given to show the effectiveness of the proposed method on the identification of influential observations.

• Communications for Statistical Applications and Methods
• /
• 제6권1호
• /
• pp.43-52
• /
• 1999

The local influence method is adapted to quadratic discriminant analysis for the identification of influential observations affecting the estimation of probability density function probabilities and log odds. The method allows a simultaneous perturbation on all observations so that it can identify multiple influential observations. The proposed method is applied to a real data set and satisfactory result is obtained.

• Journal of the Korean Statistical Society
• /
• 제29권3호
• /
• pp.337-350
• /
• 2000

This paper suggests a new diagnostic measure and a stopping rule for detecting influential observations in multiple discriminant analysis (MDA). It is developed from a Bayesian point of view using a default Bayes factor obtained from the fractional Bayes factor methodology. The Bayes factor is taken as a discriminatory information in MDA. It is shown that the effect of an observation over the discriminatory information is fully explained by the diagnostic measure. Based on the measure, we suggest a stopping rule for detecting influential observations in a given training sample. As a tool for interpreting the measure a graphical method is sued. Performance of the method is used. Performance of the method is examined through two illustrative examples.

• 응용통계연구
• /
• 제6권1호
• /
• pp.67-78
• /
• 1993

회귀진단에서 이상치와 영향을 많이 주는 측정치를 발견하는 새로운 통계량을 제안하였다. 이 제안된 통계량은 이상치를 찾는 측도와 영향추정치를 찾는 측도의 가중함으로 해석될 수 있으며, 가중치를 변화시킴으로써 이상치와 영향추정치들을 일목요연하게 찾아낼 수 있다는 장점이 있다. 씨뮬레이션을 이용하여 제안된 통계량의 분포형태를 살펴 보았다.

• Journal of the Korean Statistical Society
• /
• 제25권2호
• /
• pp.289-301
• /
• 1996

This paper suggests a new diagnostic measure for detecting influential observations in two group linear discriminant analysis(LDA). It is developed from an information theoretic point of view using the minimum discrimination information(MDI) methodology. MDI estimator of symmetric divergence by Kullback(l967) is taken as a measure of the power of discrimination in LDA. It is shown that the effect of an observation over the power of discrimination is fully explained by the diagnostic measure. Asymptotic distribution of the proposed measure is derived as a function of independent chi-squared and standard normal variables. By means of the distributions, a couple of methods are suggested for detecting the influential observations in LDA. Performance of the suggested methods are examined through a simulation study.