• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.863-874
• /
• 2021

In this paper, we propose a post-selection inference procedure for the fused lasso signal approximator (FLSA). The FLSA finds underlying sparse piecewise constant mean structure by applying total variation (TV) semi-norm as a penalty term. However, it is widely known that this convex relaxation can cause asymptotic inconsistency in change points detection. As a result, there can remain false change points even though we try to find the best subset of change points via a tuning procedure. To remove these false change points, we propose a post-selection inference for the FLSA. The proposed procedure applies a permutation test based on CUSUM statistic. Our post-selection inference procedure is an extension of the permutation test of Antoch and Hušková (2001) which deals with single change point problems, to multiple change points detection problems in combination with the FLSA. Numerical study results show that the proposed procedure is better than naïve z-tests and tests based on the limiting distribution of CUSUM statistics.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.205-205
• /
• 2019

우리나라에서 강우의 시간분포를 위해 보편적으로 사용되고 있는 방법은 Huff 4분위법으로 강우의 시간적 분포특성을 나타내는 무차원 시간분포곡선을 제시한 것으로, 강우의 지속기간을 4분위로 구분하여 각 분위의 강우량 중 가장 큰 값이 속해 있는 구간을 선택하여 그 구간의 위치에 따라 분위를 정하는 방법이다. 현재 실무에서는 Huff의 분위별 곡선에 대한 회귀식은 지속기간 전반에 걸쳐 정확도가 높은 이유로 6차식을 적용하고 있으나, 통계 모델링에서 간결함의 원리에 따라 회귀식이 간결할 필요가 있으며, 통계적 유의수준에 기초하여 회귀계수를 결정하여야 하므로 유의성 검정 방법을 통한 검정결과를 비교할 필요가 있다. 따라서 본 연구에서는 다중회귀분석 방법에 따른 회귀계수 유의성 검정결과 비교를 위하여 구미지역의 무차원 누가우량 백분율을 이용한 시간분포 회귀식을 이용하여 유의성 검정 방법인 분산분석 방법(Analysis of Variance)과 변수선택 방법(Backward Selection)의 검정 결과를 도출 및 비교하였다. 통계프로그램인 프로그래밍 R을 이용하여 변수선택 방법 중 후방제거법 함수를 이용하여 최종 회귀식을 도출하고 또한 7차 회귀식을 분산분석을 이용한 후방제거법으로 회귀계수를 제거하는 방법으로 최종 회귀식을 산정하였다. 분산분석을 이용한 후방제거법의 유의성 검정결과는 프로그래밍 R을 이용한 후방제거법의 결과와 동일한 것으로 분석되었다. 일반적으로 설계강우량의 시간분포를 위한 방법으로 사용되고 있는 Huff의 4분위 방법의 시간분포 회귀식은 회귀계수의 유의성 검정이 이루어지고 있지 않으므로 본 연구결과를 통해 설계강우량 시간분포 회귀식의 유의성 검정방법 제시 및 결과도출과정을 통해 시간분포 회귀식 산정기법으로 활용할 수 있을 것으로 사료된다.

• Journal of the Korea Society for Simulation
• /
• v.24 no.4
• /
• pp.89-96
• /
• 2015

The mediation methods suggested by Baron and Kenny, Sobel, Aroian and Goodman, have widely used to test the mediating effect. However, as there are many problems in statistical test power, as well as statistical accuracy, a bootstrapping technique has been suggested as an alternative. In this paper, we adopt the phantom variables based on the bootstrapping technique to test the mediating effect in multiple mediator model consisting of three or more mediating variables. In particular, we formulate the multiple mediator model for analyzing the relations among organizational resources, the absorption capacity as mediating variables and technology commercialization capabilities. And using the bootstrapping approach, we analyzed the mediating effect of the absorption capacity by setting of phantom variables and calculated total indirect effect size and the statistical significance. The empirical results are as follows. First, we confirmed that the bootstrapping approach and the phantom variable is the very efficient and systematic mediation method. Second, we recognized that there is a difference in the mediating characteristics of the absorption capacity depending on the resource characteristics of human resources and material resources obviously.

• The Korean Journal of Applied Statistics
• /
• v.27 no.6
• /
• pp.909-922
• /
• 2014

This paper reviews Bayesian inference with inequality constraints. It focuses on ⅰ) comparison of models with various inequality/equality constraints on parameters, ⅱ) multiple tests on equalities of parameters when parameters are under inequality constraints, ⅲ) multiple test on equalities of score parameters in models for contingency tables with ordinal categorical variables.

• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.1-13
• /
• 2019

An adaptive seamless Phase II/III clinical trial design enables a reduction in the sample size (in comparison to a conventional design) that also shortens the clinical development time. It is also very effective in clinical trials since it can have higher statistical power than Phase III alone. In this study, we use extensive simulation studies to compare several multiple hypothesis testing methods that can help select the best doses in a Phase II study along with several methods to combine p-values of the Phase II and Phase III study.

• Communications for Statistical Applications and Methods
• /
• v.19 no.1
• /
• pp.33-45
• /
• 2012

The primary interest of drug development studies is to estimate the smallest dose that shows a significant difference from the zero-dose control. The smallest dose is called the Minimum Effective dose(MED). In this paper, we suggest a nonparametric procedure to simultaneously find the MED of each group based on placements. The Monte Carlo simulation is adapted to estimate the power and the family-wise error rate(FWE) of the new procedures with those of discussed nonparametric tests to find MED.

• The Korean Journal of Applied Statistics
• /
• v.16 no.1
• /
• pp.141-150
• /
• 2003

This paper presents the multiple testing method of an autoregressive parameter in stationary AR(1) model using the usual Bayes factor. As prior distributions of parameters in each model, uniform prior and noninformative improper priors are assumed. Posterior probabilities through the usual Bayes factors are used for the model selection. Finally, to check whether these theoretical results are correct, simulated data and real data are analyzed.

• The Korean Journal of Applied Statistics
• /
• v.7 no.1
• /
• pp.103-112
• /
• 1994

An asymptotic property of the estimated eigenvalues for multivariate AR(p) process which consists of vector of nonstationary process and vector of stationary process is developed. All components of the nonstationary process are assumed to reveal random walk behavior. The asymptotic property is helpful in understanding multiple unit roots. In this paper we show the stationay part in multivariate AR(p) process does not affect the limiting distribution of estimated eigenvalues associated with the nonstationary process. A test statistic based on the ordinary least squares estimator for testing a certain number of multiple unit roots is suggested.

• Journal of the Korean Data and Information Science Society
• /
• v.12 no.1
• /
• pp.41-50
• /
• 2001

We propose the Bayesian testing for the equality of K-exponential populations means. Specially we use the intrinsic Bayesian factors suggested by Beregr and Perrichi (1996,1998) based on the noninformative priors for the parameters. And, we investigate the usefulness of the proposed Bayesian testing procedures via simulations.

• The Korean Journal of Applied Statistics
• /
• v.22 no.4
• /
• pp.759-770
• /
• 2009

A problem of separating signals from noises is considered, when they are randomly mixed in the observation. It is assumed that the noise follows a Gaussian distribution and the signal follows a Gamma distribution, thus the underlying distribution of an observation will be a mixture of Gaussian and Gamma distributions. The parameters of the mixture model will be estimated from the EM algorithm. Then the signals and noises will be classified by a fixed threshold approach based on multiple testing using positive false discovery rate and Bayes error. The proposed method is applied to a real optical emission spectroscopy data for the quantitative analysis of inclusions. A simulation is carried out to compare the performance with the existing method using 3 sigma rule.