• Wind and Structures
• /
• 제26권6호
• /
• pp.383-395
• /
• 2018

An accurate determination of wind speed distribution is the basis for an evaluation of the wind energy potential required to design a wind turbine, so it is important to estimate unknown parameters of wind speed distribution. In this paper, Gumbel distribution is used in modelling wind speed data, and alternative robust estimation methods to estimate its parameters are considered. The methodologies used to obtain the estimators of the parameters are least absolute deviation, weighted least absolute deviation, median/MAD and least median of squares. The performances of the estimators are compared with traditional estimation methods (i.e., maximum likelihood and least squares) according to bias, mean square deviation and total mean square deviation criteria using a Monte-Carlo simulation study for the data with and without outliers. The simulation results show that least median of squares and median/MAD estimators are more efficient than others for data with outliers in many cases. However, median/MAD estimator is not consistent for location parameter of Gumbel distribution in all cases. In real data application, it is firstly demonstrated that Gumbel distribution fits the daily mean wind speed data well and is also better one to model the data than Weibull distribution with respect to the root mean square error and coefficient of determination criteria. Next, the wind data modified by outliers is analysed to show the performance of the proposed estimators by using numerical and graphical methods.

• Journal of the Korean Data and Information Science Society
• /
• 제27권6호
• /
• pp.1653-1660
• /
• 2016

Smoothly clipped absolute deviation (SCAD) penalty is known to satisfy the desirable properties for penalty functions like as unbiasedness, sparsity and continuity. In this paper, we deal with the regression function estimation and variable selection based on SCAD penalized censored regression model. We use the local linear approximation and the iteratively reweighted least squares algorithm to solve SCAD penalized log likelihood function. The proposed method provides an efficient method for variable selection and regression function estimation. The generalized cross validation function is presented for the model selection. Applications of the proposed method are illustrated through the simulated and a real example.

• Journal of the Korean Data and Information Science Society
• /
• 제11권1호
• /
• pp.1-18
• /
• 2000

본 논문에서는 가장 많이 사용되는 시계열 모형중의 하나인 자기회귀모형에서 모수를 추정하는 방법으로 최소 절대 편차 추정법(least absolute deviation estimation)을 포함한 로버스트한 추정방법 (robust estimation)의 사용을 제안하고 모의 실험을 통하여 이러한 방법들을 기존의 최소 제곱 추정 방법과 예측의 관점에서 비교 검토하여 시계열 자료분석에서의 로버스트한 모수 추정 방법의 유효성을 확인해 보고자 한다.

• Journal of the Korean Statistical Society
• /
• 제33권3호
• /
• pp.323-337
• /
• 2004

This article coined a general class of estimators for various measures in normal population when some' a priori' or guessed value of standard deviation a is available in addition to sample information. The class of estimators is primarily defined for a function of standard deviation. An unbiased estimator and the minimum mean squared error estimator are worked out and the suggested class of estimators is compared with these classical estimators. Numerical computations in terms of percent relative efficiency and absolute relative bias established the merits of the proposed class of estimators especially for small samples. Simulation study confirms the excellence of the proposed class of estimators. The beauty of this article lies in estimation of various measures like standard deviation, variance, Fisher information, precision of sample mean, process capability index $C_{p}$, fourth moment about mean, mean deviation about mean etc. as particular cases of the proposed class of estimators.

• Communications for Statistical Applications and Methods
• /
• 제19권5호
• /
• pp.655-662
• /
• 2012

Classification is an important research field in pattern recognition with high-dimensional predictors. The support vector machine(SVM) is a penalized feature selector and classifier. It is based on the hinge loss function, the non-convex penalty function, and the smoothly clipped absolute deviation(SCAD) suggested by Fan and Li (2001). We developed the algorithm for the multiclass SVM with the SCAD penalty function using the local quadratic approximation. For multiclass problems we compared the performance of the SVM with the $L_1$, $L_2$ penalty functions and the developed method.

• Journal of the Korean Statistical Society
• /
• 제27권1호
• /
• pp.101-112
• /
• 1998

This paper is concerned with the asymptotic properties of the least absolute deviation estimators for the nonlinear regression model when dependent variables are subject to censoring time, and proposed the simple and practical sufficient conditions for the strong consistency and asymptotic normality of the least absolute deviation estimators in censored regression model. Some desirable asymptotic properties including the asymptotic relative efficiency of proposed model with respect to standard model are given.

• 산업경영시스템학회지
• /
• 제14권24호
• /
• pp.111-121
• /
• 1991

This paper considers a non preemtive single processor scheduling problem in which each set have the two common due dates. The objective of the problem is to minimize the weighted mean absolute deviation of job completion times about such two common due dates under the assumption that each job has a different weight. Such a job sequence is an W-shaped sequence. We propose three heuristic solution methods based on several dominance conditions. Numerical examples are presented. The performance comparison is made among three heuristic scheduling procedures.

• 한국시뮬레이션학회논문지
• /
• 제3권1호
• /
• pp.17-28
• /
• 1994

A statistical procedure is developed to estimate the relative difference between two parameters each obtained from either true model or approximate model. Double sample procedure is applied to find the additional number of simulation runs satisfying the preassigned absolute precision of the confidence interval. Two types of parameters, mean and standard deviation, are considered as the performance measures and tried to show the validity of the model by examining both queues and inventory systems. In each system it is assumed that there are three distinct means and their own standard deviations and they form the simultaneous confidence intervals but with control in the sense that the absolute precision for each confidence interval is bounded on the limits with preassigned confidence level. The results of this study may contribute to some situations, for instance, first, we need a statistical method to compare the effectiveness between two alternatives, second, we find the adquate number of replications with any level of absolute precision to avoid the unrealistic cost of running simulation models, third, we are interested in analyzing the standard deviation of the output measure, ..., etc.

• 한국전자통신학회논문지
• /
• 제9권12호
• /
• pp.1381-1388
• /
• 2014

본 논문에서는 지형의 형태 파악에 주로 이용되는 SAR(Synthetic Aperture Radar) 영상의 화질을 저해하는 주된 요소인 잡음을 제거하기 위하여 웨이블렛 변환 기반 MAD순서통계량 알고리즘을 논의한다. 효과적인 영상개선을 위하여 SAR 영상에 근사부분대역의 웨이블렛 계수에 가중평균(Weighted average)법으로 영상처리하고 상세 부분대역의 웨이블렛 계수에 중앙절대편차(MAD : Median Absolute Deviation)를 이용한 임계값을 설정하여 왜곡요소를 제거하는 방법을 제안한다. 특히 제안 방법의 임계값은 잡음과 같은 왜곡요소를 배재하고 영상의 통계량을 고려하여 설정하였다. 제안된 방법은 실시간처리를 보장하기 위하여 DSP와 FPGA를 이용한 하드웨어로 구현하였으며 Xilinx FPGA를 사용하여 실험 하였다.

• 한국안전학회지
• /
• 제22권5호
• /
• pp.84-89
• /
• 2007

폭발하한계는 가연성물질의 화재 및 폭발 위험성을 결정하는데 사용되는 중요한 연소특성치의 하나이다. 본 연구에서 에스테르 화합물에 대한 폭발하한계는 액체 열역학이론을 근거로 표준끓는점과 인화점을 이용하여 예측하였다. 그 결과, 문헌값과 예측값의 A.A.P.E.(average absolute percent error)는 8.80vo1%이고, A.A.D.(average absolute deviation)는 0.18vo1% 그리고 상관계수는 0.965로써 문헌값과 예측값은 일치하였다. 제시된 방법론 사용에 의해 다른 가연성물질의 폭발하한계 예측이 가능하다.