• Journal of the Korean Data and Information Science Society
• /
• 제24권1호
• /
• pp.125-133
• /
• 2013

이 논문에서는 호흡곤란을 주호소로 내원한 668명의 환자를 대상으로 11개 혈액검사 결과를 이용하여 퇴원여부를 결정하는 벌점 이항 로지스틱 회귀 기반 통계모형을 유도하였다. 구체적으로 $L^2$ 벌점에 근거한 능형 모형과 $L^1$ 벌점에 근거한 라소 모형을 고려하였다. 이 모형의 예측력 비교 대상으로는 일반 로지스틱 회귀의 11개 전체 변수를 사용한 모형과 변수선택된 모형이 사용되었다. 10-묶음 교차타당성 (10-fold cross-validation) 비교 결과 능형 모형의 예측력이 우수한 것으로 나타났다.

• 비교문화연구
• /
• 제43권
• /
• pp.459-483
• /
• 2016

The main objective of this paper is to test the major prediction of the Aspect Hypothesis on interlanguage narratives collected from 171 Ibibio ESL learners in a classroom setting using sets of picture stories. Aspect Hypothesis predicts that lexical aspectual classes of verbs would determine the pattern of acquisition and distribution of tense-aspect morphology at the very early stages of L2 acquisition of tense-aspect verbal morphology. That is, telic verbs would be marked with the past tense-aspect verbal morphology before atelic verbs in the interlanguage of ESL learners irrespective of their L1 background. The results of our data analyses show a significant effect from the lexical aspect on the acquisition and distribution of tense-aspect morphology with chi-square statistics of ($x^2=196.92$,df = 6, n = 1664, p = <.0001). However, the effect of the lexical aspect is shown to be more prominent among Ibibio ESL learners at higher levels of proficiency. This is contrary to the prediction regarding Aspect Hypothesis. The paper concludes that the influence of the lexical aspect on the pattern of acquisition and distribution of tense-aspect morphology may be universal but the actual point along the developmental pathway when such influence is obtainable is yet to be determined. This calls for more research into the pattern of the L2 acquisition of tense-aspect verbal morphology.

• Communications for Statistical Applications and Methods
• /
• 제22권6호
• /
• pp.543-556
• /
• 2015

We present a survey of contributions that defined the nature and extent of robust statistics for the last 50 years. From the pioneering work of Tukey, Huber, and Hampel that focused on robust location parameter estimation, we presented various generalizations of these estimation procedures that cover a wide variety of models and data analysis methods. Among these extensions, we present linear models, clustered and dependent observations, times series data, binary and discrete data, models for spatial data, nonparametric methods, and forward search methods for outliers. We also present the current interest in robust statistics and conclude with suggestions on the possible future direction of this area for statistical science.

• Journal of the Korean Statistical Society
• /
• 제29권3호
• /
• pp.285-294
• /
• 2000

In analyzing bivariate date the sample correlation coefficient is often used, and it is quite sensitive to one or few isolated cases. In this article we derive a formula for the effect of $textsc{k}$ observations on the samples correlation coefficient by the deletion method. To give a reference value for the isolated cases the asymptotic distribution fo the formula is derived. Also, we give some interpretations on several types of isolated cases and an example based on a real data set.

• Communications for Statistical Applications and Methods
• /
• 제16권1호
• /
• pp.209-215
• /
• 2009

In regression problem, the SCAD estimator proposed by Fan and Li (2001), has many desirable property such as continuity, sparsity and unbiasedness. In this paper, we extend SCAD penalized regression framework to quantile regression and hence, we propose new SCAD penalized quantile estimator on high-dimensions and also present an efficient algorithm. From the simulation and real data set, the proposed estimator performs better than quantile regression estimator with $L_1$ norm.

• Communications for Statistical Applications and Methods
• /
• 제29권3호
• /
• pp.333-351
• /
• 2022

The skew-t distribution is an attractive family of asymmetrical heavy-tailed densities that includes the normal, skew-normal and Student's-t distributions as special cases. In this work, we propose an EM-type algorithm for computing the maximum likelihood estimates for skew-t linear regression models with censored response. In contrast with previous proposals, this algorithm uses analytical expressions at the E-step, as opposed to Monte Carlo simulations. These expressions rely on formulas for the mean and variance of a truncated skew-t distribution, and can be computed using the R library MomTrunc. The standard errors, the prediction of unobserved values of the response and the log-likelihood function are obtained as a by-product. The proposed methodology is illustrated through the analyses of simulated and a real data application on Letter-Name Fluency test in Peruvian students.

• 대한수학회지
• /
• 제61권1호
• /
• pp.183-205
• /
• 2024

In this paper, for an m-dimensional (m ≥ 5) complete non-compact submanifold M immersed in an n-dimensional (n ≥ 6) simply connected Riemannian manifold N with negative sectional curvature, under suitable constraints on the squared norm of the second fundamental form of M, the norm of its weighted mean curvature vector |H_f| and the weighted real-valued function f, we can obtain: • several one-end theorems for M; • two Liouville theorems for harmonic maps from M to complete Riemannian manifolds with nonpositive sectional curvature.

• 한국연소학회지
• /
• 제18권1호
• /
• pp.13-20
• /
• 2013

Leading edge statistics are obtained by direct numerical simulation(DNS) of freely propagating incompressible and stagnating compressible turbulent premixed flames. Conditional averages of velocities in terms of reaction progress variable, c, and local flame surface density, ${\sum}^{\prime}_f$, are defined and compared through the flame brush. It holds asymptotically that $<u>_f=<S_d>_f$ and $<u>_u-<u>_b=D_t/L_w$ with the characteristic length scale of $\bar{c}$ variation, $L_w$. It also holds that $<u>_b=<u>_f$ for a freely propagating flame under no mean strain rate. The turbulent burning velocity, $S_T$, is determined by the conditional statistics at the leading edge under large activation energy.

• 응용통계연구
• /
• 제4권2호
• /
• pp.157-170
• /
• 1991

본 논문은 척도모수와 위치모수가 알려지지 않은 상황에서의 지수분포성 적합도 검정문제를 다루고 있다. 기존의 검장방법들과는 달리 누적분포 함수와 경험분포 함수 사이의 편차의 $L_1$-norm에 바탕을 두고 둔 새로운 검정방법이 제시되었으며, Monte Carlo 방법에 의하 여 검정통계량의 임계치를 구하였다. 그리고 표본의 크기가 작은 경우에 한하여 제시된 검 전통계량의 분포가 파악되었다. 한편 이 검정방법의 검정력을 기존의 검정방법들과 비교하 기 위하여 응용분야에서 흔히 사용되는 몇가지 분포형태에 대하여 검정력을 측정하였다. 그 결과, 새로운 검정방법이 보수적인 검정임에도 불구하고 다른 검정방법에 비하여 상대적으 로 검정력이 우수한 것으로 나타났다.

• Communications for Statistical Applications and Methods
• /
• 제24권5호
• /
• pp.493-505
• /
• 2017

Maximum likelihood estimation (MLE) of the generalized extreme value distribution (GEVD) is known to sometimes over-estimate the positive value of the shape parameter for the small sample size. The maximum penalized likelihood estimation (MPLE) with Beta penalty function was proposed by some researchers to overcome this problem. But the determination of the hyperparameters (HP) in Beta penalty function is still an issue. This paper presents some data adaptive methods to select the HP of Beta penalty function in the MPLE framework. The idea is to let the data tell us what HP to use. For given data, the optimal HP is obtained from the minimum distance between the MLE and MPLE. A bootstrap-based method is also proposed. These methods are compared with existing approaches. The performance evaluation experiments for GEVD by Monte Carlo simulation show that the proposed methods work well for bias and mean squared error. The methods are applied to Blackstone river data and Korean heavy rainfall data to show better performance over MLE, the method of L-moments estimator, and existing MPLEs.