# A maximum-mean-discrepancy goodness-of-fit test for censored data

arXiv: https://arxiv.org/abs/1810.04286 · HTML: https://mlsanity.com/paper/1810.04286/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Oct 2018.

## Built on by (most cited first, 9 of 9)

- [KSD Aggregated Goodness-of-fit Test](https://mlsanity.com/paper/2202.00824/) (2022)
- [Composite Goodness-of-fit Tests with Kernels](https://mlsanity.com/paper/2111.10275/) (2021)
- [Standardisation-function Kernel Stein Discrepancy: A Unifying View on Kernel Stein Discrepancy Tests for Goodness-of-fit](https://mlsanity.com/paper/2106.12105/) (2021)
- [Hypothesis testing for matched pairs with missing data by maximum mean discrepancy: An application to continuous glucose monitoring](https://mlsanity.com/paper/2206.01590/) (2022)
- [Minimax Lower Bounds of Kernel Discrepancy Estimation: MMD, HSIC, KSD](https://mlsanity.com/paper/2607.24235/) (2026)
- [A Multiple kernel testing procedure for non-proportional hazards in factorial designs](https://mlsanity.com/paper/2206.07239/) (2022)
- [Multi-Source Survival Domain Adaptation](https://mlsanity.com/paper/2212.00424/) (2022)
- [Three rates of convergence or separation via U-statistics in a dependent framework](https://mlsanity.com/paper/2106.12796/) (2021)
- [Kernel-based External Validity of Experimental Studies](https://mlsanity.com/paper/2610.13625/) (2026)

## Abstract (arXiv)

We introduce a kernel-based goodness-of-fit test for censored data, where observations may be missing in random time intervals: a common occurrence in clinical trials and industrial life-testing. The test statistic is straightforward to compute, as is the test threshold, and we establish consistency under the null. Unlike earlier approaches such as the Log-rank test, we make no assumptions as to how the data distribution might differ from the null, and our test has power against a very rich class of alternatives. In experiments, our test outperforms competing approaches for periodic and Weibull hazard functions (where risks are time dependent), and does not show the failure modes of tests that rely on user-defined features. Moreover, in cases where classical tests are provably most powerful, our test performs almost as well, while being more general.