• 응용통계연구
• /
• 제34권6호
• /
• pp.889-904
• /
• 2021

본 연구는 Kang과 Kim (2020)의 후속 연구이다. 본 연구에서는 기존 연구에서 직접 유도하지 않았던 통계량의 표본영향함수를 유도한다. 그리고 이 결과를 바탕으로 경험적 영향함수와 표본영향함수는 어떠한 관계를 가지고 있는지 이론적으로 살펴보고, 경험적 영향함수를 통해 표본영향함수를 근사시켜 추정하는 방안에 대해 생각해 본다. 또한, 임의추출한 300개의 데이터를 바탕으로 모의실험을 통해 유도한 함수와 그 관계에 대한 그 타당성도 검증한다. 모의실험 결과 t통계량으로부터 유도한 표본영향함수와 경험적 영향함수와의 관계 및 경험적 영향함수를 통한 표본영향함수의 근사 방안에 대한 타당성도 검증해 냈다. 본 연구는 경험적 영향함수를 이용한 표본영향함수의 근사에서 오차를 줄이기 위한 방안을 제안하고 그 타당성을 검증하였으며, 이를 통해 기존의 연구에서 경험적 영향함수로 표본영향함수를 바로 근사시켰던 연구 방법에 효과적인 근사 방안을 제안한 점에서 의의를 갖는다.

• Journal of the Korean Statistical Society
• /
• 제35권2호
• /
• pp.213-224
• /
• 2006

We investigate the influence of subjects or observations on regression coefficients of generalized estimating equations using the influence function and the derivative influence measures. The influence function for regression coefficients is derived and its sample versions are used for influence analysis. The derivative influence measures under certain perturbation schemes are derived. It can be seen that the influence function method and the derivative influence measures yield the same influence information. An illustrative example in longitudinal data analysis is given and we compare the results provided by the influence function method and the derivative influence measures.

• Journal of the Korean Statistical Society
• /
• 제30권3호
• /
• pp.499-509
• /
• 2001

We investigate the influence of observations on a test of additional information about discrimination using the influence function and the derivative influence measures. the influence function for the test statistic is derived and this sample versions are used for influence analysis. The derivative influence measures for the test statistic under a perturbation scheme are derived. It will be seen that the influence function method and the derivative influence measures yield the same result. Furthermore, we will derive the relationships between the influence function and the derivative influence measures when the sample size is large. an illustrative example is given and we will compare the results provided by the influence function method and the derivative influence measures.

• 응용통계연구
• /
• 제33권5호
• /
• pp.527-540
• /
• 2020

이상치에 대한 적절한 선별과 배제없이 모든 데이터를 종합적으로 분석하게 되는 경우 데이터 분석을 통해 얻은 결과의 신뢰성과 해석의 일반성에 치명적인 위협을 받을 수 있다. 따라서 데이터의 분석 과정에서 이러한 이상치를 판별하고, 이상치가 통계량, 통계적 모형에 어떠한 영향을 주는 지에 대한 분석은 매우 중요한 일이라 할 수 있다. Hampel이 영향함수를 활용하여 이상치를 판별할 수 있는 방법을 소개한 이후, 이상치를 판별하기 위한 방법론으로 영향함수가 폭넓게 활용되어 왔다. 영향함수에는 경험적 영향함수와 표본영향함수가 있으며, 경험적 영향함수를 활용해 표본영향함수를 근사 추론하여 하나의 관측값이 제거되었을 때 통계량에 미치는 영향을 예측하는 방법론이 주로 활용되었다. 본 연구에서는 표본평균, 표본분산, 표본표준편차의 표본영향함수 유도를 통해 경험적 영향함수와 표본영향함수의 차이를 살펴 본다. 또한 경험적 영향함수로 표본영향함수를 근사하는 과정에서 발생하는 오차를 줄이기 위해 경험적 영향함수의 보정으로 표본영향함수를 근사 추론하는 방법을 제안하고, 모의실험을 통해 제안한 추론 방법의 타당성을 확인한다.

• Journal of the Korean Statistical Society
• /
• 제27권4호
• /
• pp.471-483
• /
• 1998

The influence of observations on the misclassification probability in multiple discriminant analysis under the equal covariance assumption is investigated by the local influence method. Under an appropriate perturbation we can get information about influential observations and outliers by studying the curvatures and the associated direction vectors of the perturbation-formed surface of the misclassification probability. We show that the influence function method gives essentially the same information as the direction vector of the maximum slope. An illustrative example is given for the effectiveness of the local influence method.

• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.305-317
• /
• 2003

We suggest methods to measure the changes in $x^2$ statistic when a row is deleted from a two-way contingency table. The influence function is extended and the deletion method is applied. Two examples are presented and we compare the results obtained from the influence function method and the deletion method.

• Journal of applied mathematics & informatics
• /
• 제26권5_6호
• /
• pp.1133-1138
• /
• 2008

A general method of detecting outliers based on a change of likelihood by using the influence function is suggested. It can be applied to all kinds of distributions that are specified by parameters. For the multivariate normal case, specific computations are made to get the corresponding conditional influence function. A numerical example is provided for illustration.

• 한국소음진동공학회논문집
• /
• 제25권10호
• /
• pp.653-659
• /
• 2015

A new formulation of the non-dimensional dynamic influence function method, which is a type of the meshless method, is introduced to extract highly accurate eigenvalues of clamped plates with arbitrary shape. Originally, the final system matrix equation of the method, which was introduced by the author in 1999, does not have a form of algebraic eigenvalue problem unlike FEM. As the result, the non-dimensional dynamic influence function method requires an inefficient process to extract eigenvalues. To overcome this weak point, a new approach for clamped plates is proposed in the paper and the validity and accuracy is shown in verification examples.

• Communications for Statistical Applications and Methods
• /
• 제8권3호
• /
• pp.841-849
• /
• 2001

The influence in fitting an equicorrelation model is investigated using the influence function. The influence functions for the model parameters are derived and its sample versions are used for investigating the influence of observations on the estimators of the parameters. Some relationships among the sample versions are found. We will derive a measure for identifying observations that have a large influence on the test of fitting the equicorrelation model using the influence function method. An example is given for illustration.

• Communications for Statistical Applications and Methods
• /
• 제14권1호
• /
• pp.229-239
• /
• 2007

We present a diagnostic method for the quasi-likelihood estimators in generalized linear models. Since these estimators can be usually obtained by iteratively reweighted least squares which are well known to be very sensitive to unusual data, a diagnostic step is indispensable to analysis of data. We extend the local influence approach based on the maximum likelihood function to that on the quasi-likelihood function. Under several perturbation schemes local influence diagnostics are derived. An illustrative example is given and we compare the results provided by local influence and deletion.