DOI QR코드

DOI QR Code

A Comparison Study on Statistical Modeling Methods

통계모델링 방법의 비교 연구

  • Noh, Yoojeong (School of Mechanical Engineering, Pusan National University)
  • 노유정 (부산대학교 기계공학부)
  • Received : 2016.03.22
  • Accepted : 2016.05.12
  • Published : 2016.05.31

Abstract

The statistical modeling of input random variables is necessary in reliability analysis, reliability-based design optimization, and statistical validation and calibration of analysis models of mechanical systems. In statistical modeling methods, there are the Akaike Information Criterion (AIC), AIC correction (AICc), Bayesian Information Criterion, Maximum Likelihood Estimation (MLE), and Bayesian method. Those methods basically select the best fitted distribution among candidate models by calculating their likelihood function values from a given data set. The number of data or parameters in some methods are considered to identify the distribution types. On the other hand, the engineers in a real field have difficulties in selecting the statistical modeling method to obtain a statistical model of the experimental data because of a lack of knowledge of those methods. In this study, commonly used statistical modeling methods were compared using statistical simulation tests. Their advantages and disadvantages were then analyzed. In the simulation tests, various types of distribution were assumed as populations and the samples were generated randomly from them with different sample sizes. Real engineering data were used to verify each statistical modeling method.

입력 랜덤 변수(input random variable)의 통계 모델링은 기계시스템의 신뢰성 해석(reliability analysis), 신뢰성 기반 설계(reliability-based design optimization), 해석모델의 통계적 검정(validation) 및 보정(calibration)을 위해 반드시 필요하다. 대표적인 통계모델링 기법에는 Akaike Information Criterion (AIC), AIC correction (AICc), Bayesian Information Criterion, Maximum Likelihood Estimation (MLE), Bayesian 방법 등이 있다. 이러한 방법들은 기본적으로 주어진 데이터로부터 후보 모델의 우도함수값을 이용하여 후보 모델 중 가장 적합한 모델을 선택하는 방법이며, 방법에 따라 데이터 수 혹은 파라미터의 수를 고려하여 모델을 선정한다. 하지만 실제 현장에서 데이터의 통계모델링을 하는 엔지니어는 각 방법의 장단점에 대한 이해가 부족하여 어떤 방법이 정확한 방법인지 몰라 통계모델링 수행 시 어려움이 있다. 본 논문에서는 다양한 통계모델링 방법들을 비교하고 각 방법의 장단점 분석을 통해 가장 적합한 모델링 기법을 제안하고자 한다. 각 방법의 검증을 위해 다양한 모분포를 가정하고 다양한 사이즈의 샘플을 임의로 생성하여 시뮬레이션을 수행하였으며, 실제 공학 데이터를 사용하여 통계모델링 방법의 유효성을 검증하였다.

Keywords

References

  1. Y. Noh, and S. Lee "Statistical Modeling of Joint Distribution Functions for Reliability Analysis", Journal of the Korean Academia-Industrial cooperation Society, Vol. 15, No. 5, 2014, pp. 2603-2609. DOI: http://dx.doi.org/10.5762/KAIS.2014.15.5.2603
  2. Y. Noh, K. K. Choi, I. Lee, "Identification of Marginal and Joint CDFs Using the Bayesian Method for RBDO", Structural and Multidisciplinary Optimization, Vol. 40, No.1, 2010, pp.35-51. DOI: http://dx.doi.org/10.1007/s00158-009-0385-1
  3. S.-Y. Baik, B.-G., Lee, "A Study on Reliability Design of Fracture Mechanics Method Using FEM", Journal of the Korea Academia-Industrial cooperation Society, Vol. 16, No. 7, pp.4398-4404, 2015. https://doi.org/10.5762/KAIS.2015.16.7.4398
  4. H. Akaike, "A New Look at the Statistical Model Identification", IEEE Transactions on Automatic Control, Vol.19, No.6, pp. 716-723, 1974. DOI: http://dx.doi.org/10.1109/TAC.1974.1100705
  5. N. Sugiura, Further Analysts of the Data by Akaike's Information Criterion and the Finite Corrections", Communications in Statistics-Theory and Methods, Vol. 7, No. 1, 1978, pp. 13-26. DOI: http://dx.doi.org/10.1080/03610927808827599
  6. D. F. Findley, "Counter examples to Parsimony and BIC", Annals of the Institute of Statistical Mathematics, Vol. 43, No. 3, 1991, pp. 505-514. DOI: http://dx.doi.org/10.1007/BF00053369
  7. J. P. Piper, The Fatigue Properties of Formable Hot Rolled Strip Steels, Ph.D. Thesis, Univ. of Woolongong, 1993.
  8. P. Hess, D. Bruchman, I. Asskkaf, B. M. Ayyub, "Uncertainties in Material Strength, Geometric, and Load Variables", Naval Engineers Journal, Vol. 114, No. 2, 2002, pp. 139-166. DOI: http://dx.doi.org/10.1111/j.1559-3584.2002.tb00128.x

Cited by

  1. The Effects of Health Coverage Schemes on Length of Stay and Preventable Hospitalization in Seoul vol.15, pp.4, 2018, https://doi.org/10.3390/ijerph15040772
  2. A Assessment of Hydrological Drought using Tree-ring Data vol.18, pp.4, 2018, https://doi.org/10.9798/KOSHAM.2018.18.4.363