• Communications for Statistical Applications and Methods
• /
• 제11권3호
• /
• pp.643-655
• /
• 2004

L$_1$-estimator in the linear regression model is widely recognized to have superior robustness in the presence of vertical outliers. While the L$_1$-estimation procedures and algorithms have been developed quite well, less progress has been made with the hypothesis test in the multiple L$_1$-regression. This article suggests computer-intensive resampling approaches, jackknife and bootstrap methods, to estimating the variance of L$_1$-estimator and the scale parameter that are required to compute the test statistics. Monte Carlo simulation studies are performed to measure the power of tests in small samples. The simulation results indicate that bootstrap estimation method is the most powerful one when it is employed to the likelihood ratio test.

• 품질경영학회지
• /
• 제22권3호
• /
• pp.124-141
• /
• 1994

The maximum likelihood estimator (M.L.E.) and the Bayes estimators of Pr (X < Y) are derived when X and Y have a absolutely continuous bivariate exponential distribution in Block & Basu's model. The performances of M.L.E. are compared to those Bayes estimators for moderate sample size.

• 한국멀티미디어학회논문지
• /
• 제12권9호
• /
• pp.1210-1216
• /
• 2009

본 논문에서는 순환최소순위(RLR) L-필터의 수렴성을 해석하였다. RLR L-필터는 순서통계필터로서 입력의 크기순서에 따른 가중치를 필터계수로 한다. 또한 RLR L-필터는 비선형 적응 필터로서 필터계수의 갱신을 위하여 RLR 알고리즘을 이용한다. RLR 알고리즘은 로버스트 통계학의 순위추정에 기초한 비선형 적응 알고리즘이다. 본 논문에서는 가변적인 스텝 크기를 적용하여 평균 및 평균제곱의 견지에서 수렴성을 해석하였다. RLRL-필터는 잡음의 분포함수가 두꺼운 꼬리 분포인 임펄스 잡음에 가까울수록 메디안 필터의 형태로 적응하며 가우시안 잡음의 경우 평균 필터의 형태로 적응한다.

• 응용통계연구
• /
• 제23권6호
• /
• pp.1125-1145
• /
• 2010

For modeling(skewed) semicircular data, we derive a new exponential family of distributions. We extend it to the l-axial exponential family of distributions by a projection for modeling any arc of arbitrary length. It is straightforward to generate samples from the l-axial exponential family of distributions. Asymptotic result reveals that the linear exponential family of distributions can be used to approximate the l-axial exponential family of distributions. Some trigonometric moments are also derived in closed forms. The maximum likelihood estimation is adopted to estimate model parameters. Some hypotheses tests and confidence intervals are also developed. The Kolmogorov-Smirnov test is adopted for a goodness of t test of the l-axial exponential family of distributions. Samples of orientations are used to demonstrate the proposed model.

• Journal of the Korean Data and Information Science Society
• /
• 제22권3호
• /
• pp.555-563
• /
• 2011

Osteoarthritis (OA) and rheumatoid arthritis (RA) are most popular types of arthritis in Korea. This study compared health-related quality of life (HRQoL) of homedwelling people with OA and RA in Korea. Data were drawn from the Korean nationwide representative survey. Subjects were 3,352 people with arthritis over 19 years of age (2,953 OA respondents and 399 RA respondents). Good HRQoL in OA respondents was dierentiated with limitation of mobility, perceived health status, age, economic status, presence of arthralgia, gender, medical coverage, and educational level. Good HRQoL in RA respondents was dierentiated with limitation of mobility, perceived health status, economic status, educational status, and presence of arthralgia. In conclusion, HRQoL and predictors of good HRQoL among people with arthritis diers for OA or RA. These results can be of use in development of health programs and clinical interventions for community-dwelling people with arthritis.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2006년도 PROCEEDINGS OF JOINT CONFERENCEOF KDISS AND KDAS
• /
• pp.163-169
• /
• 2006

This paper deals with the comparison of parameter estimation methods in a 3-parameter Kappa distribution which is sometimes used in flood frequency analysis. The method of moment estimation(MME), L-moment estimation(L-ME), and maximum likelihood estimation(MLE) are applied to estimate three parameters. The performance of these methods are compared by Monte-carlo simulations. Especially for computing MME and L-ME, ike dimensional nonlinear equations are simplied to one dimensional equation which is calculated by the Newton-Raphson iteration under constraint. Based on the criterion of the mean squared error, the L-ME is recommended to use for small sample size $(n\leq100)$ while MLE is good for large sample size.

• Communications for Statistical Applications and Methods
• /
• 제12권2호
• /
• pp.285-294
• /
• 2005

This paper deals with the comparison of parameter estimation methods in a 3-parameter Kappa distribution which is sometimes used in flood frequency analysis. Method of moment estimation(MME), L-moment estimation(L-ME), and maximum likelihood estimation(MLE) are applied to estimate three parameters. The performance of these methods are compared by Monte-carlo simulations. Especially for computing MME and L-ME, three dimensional nonlinear equations are simplified to one dimensional equation which is calculated by the Newton-Raphson iteration under constraint. Based on the criterion of the mean squared error, L-ME (or MME) is recommended to use for small sample size( n$\le$100) while MLE is good for large sample size.

• 한국수자원학회논문집
• /
• 제32권4호
• /
• pp.437-445
• /
• 1999

비모수능형회귀분석법을 이용하여 수문응답을 추정하는 방안에 대하여 연구하였다. 응답을 추정하기 위하여 평균제곱예측오차에 대한 추정량인 CL 통계량을 최소화하는 방법을 적용하였으며 가중행렬은 전통적으로 이용도는 단위행렬과 특수한 형태인 행렬인 Laplacian 행렬을 각각 이용하여 비교하였다. 또한 추정응답의 오차분산을 추정하는 방안에 대한 검토도 실행하였다. 합성자료와 실제자료에 대한 분석 결과 가중행렬과 Laplacian 행렬을 오차분산은 편기 수정된 추정치를 이용하는 것이 좋은 결과를 보여 주었다. 본 연구에서 제시된 절차 및 방법은 수문응답 분리에 있어서 안정적이고 효율적으로 적용될 수 있을 것으로 판단된다.

• Journal of the Korean Data and Information Science Society
• /
• 제19권3호
• /
• pp.877-891
• /
• 2008

The purpose in this datamining modeling is to maximize the number of L insurance' new customer selected from the S corp.'s customers through the telemarketing. We demonstrated the superiority of this method by comparing the existing marketing method and campaign result. The used software in this analysis is SAS 9.1 and so on.

• Communications for Statistical Applications and Methods
• /
• 제24권5호
• /
• pp.443-455
• /
• 2017

In this paper, we study ARCH(${\infty}$) models with either geometrically decaying coefficients or hyperbolically decaying coefficients. Most popular autoregressive conditional heteroscedasticity (ARCH)-type models such as various modified generalized ARCH (GARCH) (p, q), fractionally integrated GARCH (FIGARCH), and hyperbolic GARCH (HYGARCH). can be expressed as one of these cases. Sufficient conditions for $L_2$-near-epoch dependent (NED) property to hold are established and the functional central limit theorems for ARCH(${\infty}$) models are proved.