• Communications for Statistical Applications and Methods
• /
• 제28권6호
• /
• pp.627-641
• /
• 2021

An asymmetric least squares estimation method has been employed to estimate linear models for percentile regression. An asymmetric maximum likelihood estimation (AMLE) has been developed for the estimation of Poisson percentile linear models. In this study, we propose generalized nonlinear percentile regression using the AMLE, and the use of the parametric bootstrap method to obtain confidence intervals for the estimates of parameters of interest and smoothing functions of estimates. We consider three conditional distributions of response variables given covariates such as normal, exponential, and Poisson for three mean functions with one linear and two nonlinear models in the simulation studies. The proposed method provides reasonable estimates and confidence interval estimates of parameters, and comparable Monte Carlo asymptotic performance along with the sample size and quantiles. We illustrate applications of the proposed method using real-life data from chemical and radiation epidemiological studies.

• 한국시뮬레이션학회논문지
• /
• 제9권4호
• /
• pp.67-75
• /
• 2000

Often system analysts are interested in the estimation of percentile for system performance. For instance, in PERT network system, the percentile that the project. Typically the control variate method is used to reduce the variability of mean response using the correlation between the response and the control variates with a little additional cost during the course of simulation. In the same spirit, we apply this method to estimate the percentile of project completion time in PERT system, and evaluate the efficiency of the controlled estimator for its percentile.1 Simulation results indicate that the controlled estimators are more effective in reducing the variances of estimators than the simple estimators, however those tend to a little underestimate the percentiles for some critical values. We need more simulation experiments to examine such a kind of bias problem. We expect this research presents a step forward in the area of variance reduction techniques of stochastic simulation.

• 한국도로학회논문집
• /
• 제16권5호
• /
• pp.165-171
• /
• 2014

PURPOSES : The objective of this study is to find Percentile Speed($V_p$) for Appraisal of Road Section with Recurring Congestion. METHODS : Percentile Speed($V_p$) is determined by correlation analysis of CSI that proposed existing literature. and CSI(Consistency Service Index) is a index that subtract service fail frequency from 100 points, and service fail is defined as traffic situation is driving less than 80kph speed. In this study, We analyzed the highest correlation percentile speed associated with CSI. This speed is chosen as a delay decision speed. In order to verifying reliability, it performed a comparison with the previous method. RESULTS : As a result, 30 percentile speed($V_{30}$) was decided as index speed for judgement of recurring congestion section, and through comparison with existing methods, we demonstrated that 30 percentile speed can be useful for judgement of recurring congestion section. CONCLUSIONS : This method to Determine recurring congestion section using the percentile speed($V_{30}$) was proposed for the first time in this paper. This method can be applicated more quickly and easily than existing method for determining of recurring delay section.

• KIEAE Journal
• /
• 제15권5호
• /
• pp.5-12
• /
• 2015

Purpose: Retention and infiltration of small and frequently-occurring rainfall by LID facilities account for a large proportion of the annual precipitation volume. Based on 4 standard facilities such as Porous Pavement, Infiltration Trench, Cylindrical Infiltration Well, Rectangular Infiltration Well by Seoul Metropolitan Handbook of the Prior Consultation for LID. The total retention volume of each facility was calculated according to the type and size. The Purpose of this study is to find out the quantitative relationship between Percentile Rainfall Event and Design Volume of Infiltration Facilities. Methode: For the estimation of Percentile Rainfall Event, Daily Precipitation of Seoul from 2005 to 2014 was sorted ascending and the distribution of percentile was estimated by PERCENTILE spreadsheet function. The managed Rainfall Depth and Percentile of each facility was calculated at the several sizes. In response to the rainwater charge volume of 5.5mm/hr by the Category "Private large site", the 3 types of facilities were planned for example. The calculated Rainfall Depth and Percentile were 54.4mm and 90% by the use of developed Calculation-Module based on the Spreadsheet program. Result: With this Module the existing Designed Infiltration volume which was introduced from Japan was simply converted to the Percentile-Rainfall-Event used in USA.

• 한국식품위생안전성학회지
• /
• 제18권3호
• /
• pp.107-112
• /
• 2003

미생물학적 위해성평가(Microbial risk assessment: MRA)에서 노출평가시 확률적 접근방법은 식중독 발생에 관련도 많은 위해 미생물의 다양성과 변이성 그리고 불확실성에 대한 분석이 가능하고 과거의 접근방법인 point estiamte보다 훨씬 더 현실성이 반영된 결과를 제공할 수 있어 현재 MRA의 노출평가에서 가장 효과적인 방법으로 인정되고 있다. 본 연구는 MRA 방법론 중 노출평가에서 이용될 수 있는 수학적 확률분포 모델을 이용하여 위해 미생물의 발생수준을 추정하는 방법론을 제시하고, 이를 바탕으로 국내 식육에서의 Listeria monocytogenes에 대한 오염수준을 추정하였으며, 그 결과 국내 식육에서의 L. monocytogenes의 오염수준은 평균($50^{th}$ percentile)-4.08 Log CFU/g이며, 최소($5^{th}$ percentile) -4.88 Log CFU/g, 최대 ($95^{th}$ percentile) -3.56 Log CFU/g의 범위를 갖는 것으로 추정되었다.

• 대한인간공학회지
• /
• 제7권1호
• /
• pp.5-12
• /
• 1988

Most designers of equipment, or of a workplace layout, frequently choose 5 percentile female and 95 percentile male dimensions as design limits. If the design is not an one-dimensional problem, it is difficult to choose '5 percentile female' or '95 percentile male'. Hence it is important to consider the interactions or relationships between body dimensions. This paper deals with correlation analysis, factor analysis and partial correlation analysis using data above age 20 from the national anthropometric survey in Korea, 1986. Besides, a model which represent the extremes of body size of Korean has been developed in this paper.

• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.551-562
• /
• 1995

We consider the iterated bootstrap method in regression model with heterogeneous error variances. The iterated wild bootstrap confidence intervla of the mean response is considered. It is shown that the iterated wild bootstrap confidence interval has coverage error of order $n^{-1}$ wheresa percentile method interval has an error of order $n^{-1/2}$. The simulation results reveal that the iterated bootstrap method calibrates the coverage error of percentile method interval successfully even for the small sample size.

• 정보처리학회논문지D
• /
• 제10D권6호
• /
• pp.1025-1032
• /
• 2003

백분위수는 시뮬레이션 결과의 전체적인 성향을 파악하는데 아주 유용한 측정 기법 중의 하나이다. 그러나, 시뮬레이션으로 수집된 데이터들에 대한 평균이나 표준편차와는 달리 백분위수를 추정하기 위해서는 모든 관측된 데이터들을 저장해야 만 한다, 왜냐하면 백분위수의 추정을 위해서는 관측된 모든 데이를 분류하여 오른차순으로 정렬하는 등 여러 단계의 처리과정이 필요하기 때문이다. 따라서, 백분위수 추정을 위해서는 관측된 모든 데이터를 저장하기 위한 대용량의 저장장치와 정렬을 위한 계산시간 (O($nlog_{2}n$))이 요구된다. 이러한 문제점을 해결하기 위한 여러 백분위수 추정 기법들이 제안되었으나 고정된 샘플 크기의 시뮬레이선(fixed sample size simulation) 을 수행할 경우에만 적용 가능하다. [11, 12, 21]. 본 논문에서는 3가지 백분위수 추정 기법(linear PE, batching PE, spectral $P^2$ PE) 을 순차적인 안정상태 시뮬레이션(sequential steady-state simulation) 에 적용하여 연구하였다. 또한, 3가지의 백분위수 추정 기법들에 대해 coverage 분석을 수행한 결과를 제시하였다.

• 산업경영시스템학회지
• /
• 제25권6호
• /
• pp.17-22
• /
• 2002

In this paper, We consider some generalization of these five basic indices to cover non-normal distribution. The proposed generalizations are compared with the five basic indices. The results show that the proposed generalizations are more accurate than those basic indices and other generalization in measuring process capability. We compared an estimation methods by Clements with based on sample percentiles WVM to calculate the proposed generalization as an example The results indicated that Clements method is more accurate than percentile method, WVM in measuring process capability But the calculations of percentile method are easy to understand, straightforward to apply, and show be valuable used for applications.

• Communications for Statistical Applications and Methods
• /
• 제6권3호
• /
• pp.977-994
• /
• 1999

Five equating methos for elective subjects of College Scholastic Ability Test were analyzed under a common-items nonequivalent groups design using a real data set of 110 thousand examinees. Five methods were (a)two-stage linear equating (b) two-stage equi-percentile equating (c) Tucker equating (d) Frequency estimation equating and (e)Braun-Holland equating. The results indicated that Frequency estimation equating fits well and two-stage linear equating produces most different equating results from the Frequency estimation equating.