A nonparametric test for comparing two samples where all observations are either left- or right-censored

Area of Science:

• Biostatistics
• Survival Analysis
• Statistical Methods

Background:

• Assessing individual status once leads to censored time-to-event data.
• Observed times can be either left-censored or right-censored.
• Accurate estimation is crucial for understanding event occurrences.

Purpose of the Study:

• To review nonparametric estimation techniques for time-to-event distributions.
• To propose a novel nonparametric two-sample test for survival data.
• To evaluate the performance of the proposed test.

Main Methods:

• Review of existing nonparametric estimation methods for censored data.
• Development of a new nonparametric two-sample test.
• Performance evaluation using Monte Carlo simulations.
• Illustration with a numerical example.

Main Results:

• Nonparametric estimation techniques for censored data were systematically reviewed.
• A new nonparametric two-sample test was developed and presented.
• Simulations demonstrated the test's performance characteristics.
• The test's applicability was shown through a numerical example.

Conclusions:

• The proposed nonparametric two-sample test is a viable method for analyzing time-to-event data with censoring.
• The test provides a valuable tool for comparing event distributions in two groups.
• Further research can explore extensions and applications of this test.