Weighted log rank statistics for comparing two distributions

Moreau, Thierry; Maccario, Jean; Lellouch, J; Huber, Catherine

doi:10.1093/biomet/79.1.195

Cited by 29 publications

(17 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…11. In contrast to the classical setting, for which Moreau et al [13] state that LR AW performs better than LR and PPW, LR AW has the least power for all of the scenarios in group sequential setting. Therefore, usage of LR AW in group sequential setting would be misleading.…”

Section: Resultsmentioning

confidence: 74%

“…The last considered weighted test statistic, given by Moreau [13], is LR AW . It includes the case of two Weibull distributions differing in shape parameters.…”

Section: Weighted Testsmentioning

confidence: 99%

“…Their GSTs are illustrated over a reviewed general form. In addition, a new test, which is denoted by LR AW (asymptotic weighted LR test) and proposed by Moreau et al [13], is briefly reviewed. The group sequential testing procedure for this test was carried out over the general form.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A simulation study on power comparisons for group sequential tests of non-parametric statistics

Demirhan¹,

Demirhan

Bacanli

2009

Journal of Statistical Computation and Simulation

View full text Add to dashboard Cite

In this article, a group sequential test (GST) of non-parametric statistics for survival data is briefly reviewed. An asymptotic joint distribution of the test statistics, obtained after each interim analysis, is given to illustrate the applicability of the critical values of the GST procedures. It should be noted that censored observations are generally seen in survival data. Therefore, if one makes power calculations irrespective of censoring, reliable results may not be achieved, due to the lack of information about the censoring structure. A wide simulation study, covering different censoring rates and tied observations, is conducted to make the power comparisons under various scenarios. The simulation results are interpreted and compared with the results obtained by using power analysis and sample size (PASS) software.

show abstract

Section: Resultsmentioning

confidence: 74%

“…The last considered weighted test statistic, given by Moreau [13], is LR AW . It includes the case of two Weibull distributions differing in shape parameters.…”

Section: Weighted Testsmentioning

confidence: 99%

See 1 more Smart Citation

A simulation study on power comparisons for group sequential tests of non-parametric statistics

Demirhan¹,

Demirhan

Bacanli

2009

Journal of Statistical Computation and Simulation

View full text Add to dashboard Cite

show abstract

“…is the Prentice estimation of the survival function that appears in Moreau et al (1992). We observe that the weighted and score tests do not use the same variancecovariance matrix, so this equivalence must be understood in the sense that S ¼ U although they are not the same test, due to: X 2 S 6 ¼ X 2 U .…”

Section: Four New Testsmentioning

confidence: 97%

Survival Tests forr Groups

Letón¹,

Zuluaga²

2002

Biom. J.

View full text Add to dashboard Cite

“…These methods can be roughly classified into three groups: (i) Omnibus tests such as the modified Kolmogorov-Smirnov test (Fleming et al 1980), the Renyi-type test (Gill 1980) and Liu et al (2007)'s test; (ii) Weighted log-rank tests whose weights change signs before and after a potential crossing point (i.e., Mantel and Stablein 1988;Moreau et al 1992;Qiu and Sheng 2008); and (iii) Methods based on explicitly modeling the crossing structure of the hazard rates (Anderson and Senthilselvan 1982;Breslow et al 1984;Liu et al 2007;Bagdonavičius et al 2012). Comparing the above three groups of approaches, we expect the second and third groups to be more powerful in testing differences between two crossing hazard rates, because they are designed specifically for testing the alternatives of crossing hazard rates, instead of some more general alternatives that are aimed at by the first group of methods (Liu et al 2007).…”

Section: Introductionmentioning

confidence: 99%

Partitioned log-rank tests for the overall homogeneity of hazard rate functions

Liu

Yin

2016

Lifetime Data Anal

View full text Add to dashboard Cite

In survival analysis, it is routine to test equality of two survival curves, which is often conducted by using the log-rank test. Although it is optimal under the proportional hazards assumption, the log-rank test is known to have little power when the survival or hazard functions cross. To test the overall homogeneity of hazard rate functions, we propose a group of partitioned log-rank tests. By partitioning the time axis and taking the supremum of the sum of two partitioned log-rank statistics over different partitioning points, the proposed test gains enormous power for cases with crossing hazards. On the other hand, when the hazards are indeed proportional, our test still maintains high power close to that of the optimal log-rank test. Extensive simulation studies are conducted to compare the proposed test with existing methods, and three real data examples are used to illustrate the commonality of crossing hazards and the advantages of the partitioned log-rank tests.

show abstract

Weighted log rank statistics for comparing two distributions

Cited by 29 publications

References 8 publications

A simulation study on power comparisons for group sequential tests of non-parametric statistics

A simulation study on power comparisons for group sequential tests of non-parametric statistics

Survival Tests forr Groups

Partitioned log-rank tests for the overall homogeneity of hazard rate functions

Contact Info

Product

Resources

About