Journal of the Korean Data and Information Science Society

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

DOI QR Code

A comparison of the statistical methods for testing the equality of crossing survival functions

교차하는 두 생존함수의 동일성 검정법에 관한 비교연구

  • Received : 2015.01.27
  • Accepted : 2015.04.06
  • Published : 2015.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

Log-rank is widely used for testing equality of two survival functions, and this method is efficient only under the proportional hazard assumption. However, crossing survival functions are common in practice. Therefore, many approaches have been suggested to test equality of them. This study considered several methods; Renyi type test, modified Kolmogorov-Smirnov and Cramer-von Mises test, and weighted Log-rank test, which can be applied when the survival functions cross, and simulated power of those methods. Based on the simulation results, we provide the useful information to choose a suitable approach in a given situation.

두 집단의 생존분포의 비교를 위한 방법으로는 로그-순위 검정법이 널리 쓰이고 있다. 그러나 이는 두 집단의 생존분포 간에 비례위험모형 가정이 성립하는 경우에 적합한 것으로써, 두 집단의 생존분포가 교차하는 상황 하에서는 해당 검정법의 유효성을 장담할 수 없다. 그러나 두 집단의 생존분포가 교차하는 상황은 빈번히 발생하며, 이 같은 상황에서 두 생존분포의 동일성 검정을 위한 여러 연구가 진행되어왔다. 본 논문에서는 두 집단 간 생존분포가 교차하는 상황 하에서의 동일성 검정을 위한 방법들을 고려한다. 나아가 교차하는 상황을 위치에 따라 세분화하고 모의실험을 통해 각 상황에서 다양한 검정법들의 검정력을 비교하였으며, 그 결과를 토대로 주어진 상황에서 적절한 방법의 선택에 유용한 정보를 제공하고자 한다.

Keywords

References

  1. Aalen, O. O. (1976). Nonparametric inference in connection with multiple decrement models. Scandinavian Journal of Statistics, 3, 15-27.
  2. Aalen, O. O. (1978). Nonparametric estimation of partial transition probabilities in multiple decrement models. The Annals of Statistics, 6, 534-545. https://doi.org/10.1214/aos/1176344198
  3. Breslow, N. and Crowley, J. (1974). A large sample study of the life table and product limit estimates under random censorship. The Annals of Statistics, 2, 437-453. https://doi.org/10.1214/aos/1176342705
  4. Kitabo, C. A. and Kim, J. T. (2014). Survival analysis of bank loan repayment rate for customers of Hawassa commercial bank of Ethiopaia. Journal of the Korean Data & Information Science Society, 25, 1591-1598. https://doi.org/10.7465/jkdi.2014.25.6.1591
  5. Davis, M. and Xie, S. X. (2011). Caution: Hazards crossing! using the renyi test statsitic in survival analysis, PharmaSUG 2011, Tennessee.
  6. Fleming, T. R. and Harrington, D. P. (1991). Counting processes and survival analysis, Wiley, New York.
  7. Gill, R. D. (1980) Censoring and stochastic integrals, The Mathematical Centre, Netherlands.
  8. Jung, M. N. and Lee, J. W. (1998). A comparison of the statistical methods for testing the equality of two survival distributions. The Korean journal of Applied statistics, 11, 113-127
  9. Klein, J. P. and Moeschberger, M. L. (2003). Survival analysis: Techniques for censored and truncated data, 2nd Ed., Springer, Berlin.
  10. Lee, J. W. (1996). Some versatile tests based on the simultaneous use of weighted log-rank statistics. Biometrics, 52, 721-725. https://doi.org/10.2307/2532911
  11. Lee, W. K., Song, M. U. and Jeong, S. H. (2011). Estimation of conditional mean residual life function with random censored data. Journal of the Korean Data & Information Science Society, 22, 89-97.
  12. Logan, B. R., Klein, J. P. and Zhang, M. J. (2008). Comparing treatment in the Presence of crossing survival curves: An application to bone marrow transplantation. Biometrics, 64, 733-740. https://doi.org/10.1111/j.1541-0420.2007.00975.x
  13. Nelson, W. B. (1969). Hazard plotting for incomplete failure data. Journal of Quality Technology, 1, 27-52. https://doi.org/10.1080/00224065.1969.11980344
  14. Park, J. K., Oh, K. H. and Kim, M. S. (2010). A default-rate comparison of the construction and other industries using survival method. Journal of the Korean Data & Information Science Society, 21, 747-756.
  15. Qui, P. and Sheng, J. (2008). A two-stage procedure for comparing hazard rate functions. Journal of the Royal Statistical Society B, 70, 101-208.
  16. Schumacher, M. (1984). Two-sample tests of cramer-von mises- and kolmogorov-smirnov-type for randomly censored data. International Statistical Review, 52, 263-281. https://doi.org/10.2307/1403046