# A Conditional Independence Test in the Presence of Discretization

arXiv: https://arxiv.org/abs/2404.17644 · HTML: https://mlsanity.com/paper/2404.17644/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 44 references that are in the map.
- First posted on arXiv 26 Apr 2024.

## Builds on (most cited first, 5 of 5)

- [Approximate Kernel-based Conditional Independence Tests for Fast Non-Parametric Causal Discovery](https://mlsanity.com/paper/1702.03877/) (2017)
- [Model-Powered Conditional Independence Test](https://mlsanity.com/paper/1709.06138/) (2017)
- [Federated Causal Discovery from Heterogeneous Data](https://mlsanity.com/paper/2402.13241/) (2024)
- [Asymptotic normality and optimalities in estimation of large Gaussian graphical models](https://mlsanity.com/paper/1309.6024/) (2013)
- [High Dimensional Semiparametric Latent Graphical Model for Mixed Data](https://mlsanity.com/paper/1404.7236/) (2014)

## Built on by (most cited first, 3 of 3)

- [Permutation-Based Rank Test in the Presence of Discretization and Application in Causal Discovery with Mixed Data](https://mlsanity.com/paper/2501.18990/) (2025)
- [A Sample Efficient Conditional Independence Test in the Presence of Discretization](https://mlsanity.com/paper/2506.08747/) (2025)
- [Conditional Independence Tests for Constraint-Based Causal Discovery: A Survey](https://mlsanity.com/paper/2608.11156/) (2026)

## Abstract (arXiv)

Testing conditional independence has many applications, such as in Bayesian network learning and causal discovery. Different test methods have been proposed. However, existing methods generally can not work when only discretized observations are available. Specifically, consider X1, tildeX2 and X3 are observed variables, where tildeX2 is a discretization of latent variables X2. Applying existing test methods to the observations of X1, tildeX2 and X3 can lead to a false conclusion about the underlying conditional independence of variables X1, X2 and X3. Motivated by this, we propose a conditional independence test specifically designed to accommodate the presence of such discretization. To achieve this, we design the bridge equations to recover the parameter reflecting the statistical information of the underlying latent continuous variables. An appropriate test statistic and its asymptotic distribution under the null hypothesis of conditional independence have also been derived. Both theoretical results and empirical validation have been provided, demonstrating the effectiveness of our test methods.