Journal of the Korean Data and Information Science Society

The Korean Data and Information Science Society (한국데이터정보과학회)
권호 표지
DOI QR코드

DOI QR Code

ROC curve and AUC for linear growth models

선형성장모형에 대한 ROC 곡선과 AUC

  • Received : 2015.09.10
  • Accepted : 2015.11.09
  • Published : 2015.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

Consider the linear growth models for longitudinal data analysis. Several kind of linear growth models are selected such as time-effect and random-effect models as well as a dummy variable included model. In this work, simulation data are generated with normality assumption, and both binormal ROC curve and AUC are obtained and compared for various linear growth models. It is found that ROC curves have different shapes and AUC increase slowly, as values of the covariance increase and the time passes for random-effect models. On the other hand, AUC increases very fast as values of covariance decrease. When the covariance has positive value, we explored that the variances of random-effect models increase and the increment of AUC is smaller than that of AUC for time-effect models. And the increment of AUC for time-effect models is larger than the increment for random-effect models.

경시적자료의 분석으로 선형성장모형을 고려한다. 시간효과를 고려하는 모형과 임의효과를 추가하는 모형 그리고 가변수가 추가된 모형을 설정한다. 본 연구는 정규분포로 가정한 다양한 자료를 생성하고, 다양한 선형성장모형에 대하여 binormal ROC 곡선과 AUC 통계량을 여러 시점에서 구하여 비교 분석하였다. 공분산의 크기가 증가할수록 그리고 시간이 경과할수록 ROC 곡선은 다른 형태로 나타나며 AUC 값은 서서히 증가한다. 반대로 공분산이 작아질수록 시간이 경과함에 따라 AUC의 증가폭이 커진다. 임의효과모형에서 공분산이 양인 경우에 시간이 경과할수록 임의효과모형의 분산이 증가하며 AUC의 증가량은 시간효과모형의 AUC의 증가량보다 작다. 그리고 시간효과모형의 AUC의 증가량보다 임의효과모형의 증가량이 더 크다는 것을 탐색하였다.

Keywords

References

  1. Ashenfelter, O., Levine, B. P. and Zimmerman, J. D. (2003). Statistics and econometrics: Methods and applications, John Wiley & Sons, New Jersey.
  2. Bloom, L., Persson, L. A. and Dahlquist, G. (1992). A high linear growth is associated with an increased risk of childhood diabetes mellitus. Diabetologia, 35, 528-533. https://doi.org/10.1007/BF00400480
  3. Etzioni, R., Pepe, M. S., Longton, G., Hu, C. and Goodman, G. (1999). Incorporating the time dimension in receiver operating characteristic curve: A case study of prostate cancer. Medical Decision Making, 19, 242. https://doi.org/10.1177/0272989X9901900303
  4. Hong, C. S., Kim, G. C. and Jeong, J. A. (2012). Bivariate ROC curve. Communications of The Korean Statistical Society, 19, 277-286. https://doi.org/10.5351/CKSS.2012.19.2.277
  5. Hong, C. S., Jung, E. S. and Jung, D. G. (2013). Standard criterion of VUS for ROC surface. The Korean Journal of Applied Statistics, 26, 977-985. https://doi.org/10.5351/KJAS.2013.26.6.977
  6. Hong, C. S. and Jung, D. G. (2014). Standard criterion of hypervolume under the ROC manifold. Journal of the Korean Data & Information Science Society, 25, 473-483. https://doi.org/10.7465/jkdi.2014.25.3.473
  7. Jeon, J. Y. and Lee, K. B. (2014). Review and discussion of marginalized random effects models. Journal of the Korean Data & Information Science Society, 25, 1263-1272. https://doi.org/10.7465/jkdi.2014.25.6.1263
  8. Johnston, J. and Dinardo, J. (2000). Econometric methods. Econometric Theory, 16, 139-142
  9. Pearson, J. D., Luderer, A. A., Metter, E. J., Partin, A. W., Chan, D. W., Fozard, J. L. and Carter, H. B. (1996). Longitudinal analysis of series measurements of free and total psa among men with and without prostatic cancer. Elsevier Science, 48, 4-9.
  10. Pepe, M. S. (1997). A regression modelling framework for receiver operating characteristic curves in medical diagnositic testing. Biometrika, 84, 595-608. https://doi.org/10.1093/biomet/84.3.595
  11. Pepe, M. S. (1998). Three approaches to regression analysis of receiver operating characteristic curves for continuous test results. Biometrics, 54, 124-135. https://doi.org/10.2307/2534001
  12. Pepe, M. S. (2000). An interpretation for the ROC curve and inference using glm procedure. Biometrics, 56, 352-359. https://doi.org/10.1111/j.0006-341X.2000.00352.x
  13. Park, H. J., Sang, K. A. and Kang, J. Y. (2008). Effect of private tutoring on middle school students' achievement. Journal of Educational Evaluation, 21, 107-127.
  14. Tosteson, A. N. and Begg, C. B. (1998). A general regression methodology for ROC curve estimation. Medical Decision Making, 8, 204.

Cited by

  1. 계수적 반응을 갖는 종양 억제 혼합물 실험에서 모형 비교 vol.28, pp.5, 2017, https://doi.org/10.7465/jkdi.2017.28.5.1021