• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.868-871
• /
• 2021

기후, 경영, 경제 등 여러 분야의 회귀분석에서 설명변수가 반응변수에 일정 시차를 두고 영향을 미치는 경우들이 많다. 하지만 지금까지 대부분의 회귀분석은 설명변수가 반응변수에 즉각적으로 영향을 미치는 경우만을 가정하고 있으며, 설명변수와 반응변수 간에 존재하는 시차를 탐색하는 연구는 거의 이루어지지 않았다. 그러나 보다 정확한 회귀분석을 위해서는 설명변수와 반응변수 간에 존재하는 시차를 파악하는 것이 중요하다. 본 논문은 회귀분석 데이터가 주어졌을 때 설명변수와 반응변수 간에 존재하는 시차를 파악하는 딥러닝 모델을 제안한다. 제안하는 딥러닝 모델은 설명변수의 과거 값들 중 어떤 값이 현재 반응변수에 가장 큰 영향을 미치는지를 노드 간 가중치로 표현하고, 회귀모델의 오차를 최소화하는 가중치를 탐색한다. 훈련이 끝나면 이 가중치들을 사용하여 각 설명변수와 반응변수 간에 존재하는 시차를 파악한다. 실험을 통해 제안 방법은 시차를 고려하지 않는 기존 회귀모델에 비해 시차까지 고려함으로써 오차가 1/100 수준에 불과한 더 정확한 회귀모델을 찾을 수 있음을 확인하였다.

• Journal of the Korea Society for Simulation
• /
• v.21 no.3
• /
• pp.71-78
• /
• 2012

This research discusses the application of the control variables to achieve a more precise estimation for the target response in queueing network simulation. The efficiency of control variable method in estimating the response depends upon how we choose a set of control variables strongly correlated with the response and how we construct a function of selected control variables. For a class of queuing network simulations, the random variables that drive the simulation are basically the service-time and routing probability random variables. Most of applications of control variable method focus on utilization of the service time random variables for constructing a controlled estimator. This research attempts to suggest a controlled estimator which uses these two kinds of random variables and explore the efficiency of these estimators in estimating the reponses for computer network system. Simulation experiments on this model show the promising results for application of routing probability control variables. We consider the applications of the routing probability control variables to various simulation models and combined control variables using information of service time and routing probability together in constructing a control variable as future researches.

• The Korean Journal of Applied Statistics
• /
• v.34 no.4
• /
• pp.659-669
• /
• 2021

Multivariate regression often appears in longitudinal or functional data analysis. Since multivariate regression involves multi-dimensional response variables, it is more strongly affected by the so-called curse of dimension that univariate regression. To overcome this issue, Yoo (2018) and Yoo (2019a) proposed three model-based response dimension reduction methodologies. According to various numerical studies in Yoo (2019a), the default method suggested in Yoo (2019a) is least sensitive to the simulated models, but it is not the best one. To release this issue, the paper proposes an selection algorithm by comparing the other two methods with the default one. This approach is called principal selected response reduction. Various simulation studies show that the proposed method provides more accurate estimation results than the default one by Yoo (2019a), and it confirms practical and empirical usefulness of the propose method over the default one by Yoo (2019a).

• The Korean Journal of Applied Statistics
• /
• v.14 no.2
• /
• pp.321-332
• /
• 2001

본 논문에서는 반응변수가 두 가지의 값을 갖는 회귀분석에 적용할 수 있는 사영추적회귀를 고려하였다. 회귀모형에 필요한 설명변수들의 선형결합이 하나이고 연결함수의 형태를 사전에 알지 못한다는 가정하에서 모의담금질 기법을 이용하여 모형에 필요한 선형결합을 찾는 알고리즘을 제시하였다. 이진 반응변수의 경우에는 평활모수의 값에 따라 잔차이탈도함수의 반응표면이 단봉의 형태를 갖지 않는 경우가 있어 비동질적 마코프체인을 이용한 모의담금질 기법을 적용하면 효율적으로 선형결합을 탐색할 수 있다.

• Journal of Korean Society of Environmental Engineers
• /
• v.29 no.12
• /
• pp.1366-1370
• /
• 2007

For the purpose of finding how the annoyance response to aircraft noise is affected by non-noise variables, the questionnaire survey is conducted around the Gimpo International Airport in Seoul, Korea. The non-noise variables used in this research are divided into two categories; demographic and attitudinal variables. The result of the survey suggests that aircraft noise annoyance is not affected to an important extent by other noise sources(e.g., road traffic noise and community noise etc.) and the demographic variables (sex, age, education, occupation, dwelling type and length of residence). It has been found that it is affected to an important extent by the attitudinal variables such as complaints.

• Journal of Korea Society of Industrial Information Systems
• /
• v.15 no.5
• /
• pp.179-185
• /
• 2010

In a sensitivity analysis, an uncertainty importance measure is often used to assess how much uncertainty of an output is attributable to the uncertainty of an input, and thus, to identify those inputs whose uncertainties need to be reduced to effectively reduce the uncertainty of output. A function is called monotonic if the output is either increasing or decreasing with respect to any of the inputs. In this paper, for a monotonic function, we propose a method for evaluating the measure which assesses the expected percentage reduction in the variance of output due to ascertaining the value of input. The proposed method can be applied to the case that the output is expressed as linear and nonlinear monotonic functions of inputs, and that the input follows symmetric and asymmetric distributions. In addition, the proposed method provides a stable uncertainty importance of each input by discretizing the distribution of input to the discrete distribution. However, the proposed method is computationally demanding since it is based on Monte Carlo simulation.

• The Korean Journal of Applied Statistics
• /
• v.21 no.1
• /
• pp.35-51
• /
• 2008

Comparative studies in Phase III clinical trials quite often involve two or more equally important endpoints, and one cannot select primary endpoint from them. O'Brien(1984) proposed for continuous endpoints the OLS and GLS statistics as milti-variate test statistics. Pocock et al. (1987) mentioned the possibility of analyzing a mixture of data types, such as quantitative, binary and survival data types, with the OLS and GLS statistics, but the authors did not explore problems in combining several endpoints of different types. Furthermore, they did not perform a simulation study to assess the efficiencies of the OLS and GLS statistics for endpoints of a mixture of data types. In this paper, we propose the combining methods of correlated P-values for the analysis of multiple endpoints, and compare the efficiencies of this method with those of OLS and GLS statistics for a mixture of data types with a simulation study. Among the several methods of combining P-values that are more advantageous than combining of OLS and GLS statistics, method B maintains nominal significance levels and is more efficient, while method F and G have type I error rates that are larger than the specified significance levels, which might occasionally lead to a wrong conclusion.

• Journal of the Korea Society for Simulation
• /
• v.19 no.1
• /
• pp.133-141
• /
• 2010

This research suggests a unified scheme for using stratified sampling and control variates method to improve the efficiency of estimation for parameters in simulation experiments. We utilize standardized concomitant variables defined during the course of simulation runs. We first use these concomitant variables to counteract the unknown error of response by the method of control variates, then use a concomitant variable not used in the controlled response and stratify the response into appropriate strata to reduce the variation of controlled response additionally. In case that the covariance between the response and a set of control variates is known, we identify the simulation efficiency of suggested method using control variates and stratified sampling. We conjecture the simulation efficiency of this method is better than that achieved by separated application of either control variates or stratified sampling in a simulation experiments. We investigate such an efficiency gain through simulation on a selected model.

• KSBB Journal
• /
• v.5 no.2
• /
• pp.133-139
• /
• 1990

A simple and convenient method was introduced to determine the kinetic parameters for various enzymatic reaction kinetics. The method based on integrated formular can be applied to the parameter estimations from a single experiment. A modified three-parameter model was applied for the parameter estimation in reversible reaction and the equilibrium substrate concentration could be also estimated. It is possible to identify the enzymatic reaction pattern by inspecting the parameter values and the square of the correlation coefficient.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2004.05a
• /
• pp.41-47
• /
• 2004

중심합성계획(central composite design: ccd)은 반응 표면이 곡면적인 특성을 나타낼때 반응 공간을 추정하기 위해 사용되는 실험계획이다. 반응공간이 2차 회귀모형으로 나타나는 경우에 반응곡면의 변화량을 알기 위해서는 변수의 수준이 3이상이 되어야하는데 ccd는 적은 횟수의 실험으로 곡면을 효과적으로 추정하기 위해 2$^{k}$ 요인실험에 추가적으로 중심점(central point)과 축점(axial point)을 표본점에 포함시키는 계획이다. 본 연구에서는 시뮬레이션 실험에서 반응변수가 2차 회귀모형으로 근사되는 경우에 cod를 이용하여 관심 성과치의 반응표면을 추정하고자 한다. 일반적인 실험에서와는 달리 시뮬레이션 실험에서는 두개의 표본점(인자 수준의 조합)에서 분석자가 공통 난수계열(common random number series)을 부여하여 시뮬레이션 시스템 요소의 변화과정을 유사하게 통제할 수 있다. 일반적으로 공통난수법(common random number method)에 의해 얻어지는 두 표본점에서의 반응변수는 서로 양의 상관관계를 가지며 대조 난수(antithetic random number)에 의한 두 반응변수는 음의 상관성을 가지는 것으로 알려졌다. 본 연구는 ccd의 표본점에 공통난수와 대조난수 법을 이용하여 회귀모형의 파라미터를 효과적으로 추정하는 방법을 조사하고 이를 (s, S) 재고관리 모형에 적용하여 그 효율성을 평가하고자 한다.