Incremental Causal Effect for Time to Treatment Initialization
Authors: Andrew Ying, Zhichen Zhao, Ronghui Xu
ICLR 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We illustrate our approach via simulation, and apply it to a rheumatoid arthritis study to evaluate the incremental effect of time to start methotrexate on joint pain. In this section, we investigate the finite-sample performance of our estimator. We examine the performance of the IPW estimator ˆψ(θ) and its variance estimate when θ(t, l) 1/3, 1/2.5, 1/2, 1/1.5, 1.5, 2, 2.5, 3 by reporting biases, percent biases (%Bias), empirical standard errors (SEE)... |
| Researcher Affiliation | Academia | Andrew Ying Irvine, CA 92606, USA EMAIL Zhichen Zhao Department of Mathematics University of California, San Diego La Jolla, CA 92093, USA EMAIL Ronghui Xu Herbert Wertheim School of Public Health & Human Longevity Science, Department of Mathematics and Halıcıo glu Data Science Institute University of California, San Diego La Jolla, CA 92093, USA EMAIL |
| Pseudocode | No | The paper describes its estimation framework using inverse probability weighting and presents formulas for the estimator, but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement about releasing source code or a link to a code repository for the described methodology. |
| Open Datasets | Yes | We illustrate our framework by estimating incremental causal effects of the anti-rheumatic therapy Methotrexate (MTX) among patients with rheumatoid arthritis, using data from Choi et al. (2002). |
| Dataset Splits | No | The paper uses simulated data and a clinical dataset, but it does not specify any training, testing, or validation splits for either. For the clinical data, it only mentions inclusion criteria for selecting patients from a larger dataset. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used to perform the simulations or apply the framework to the rheumatoid arthritis study. |
| Software Dependencies | No | The paper mentions using a 'Cox proportional hazard model' and 'multiplier bootstrap' as methodologies but does not specify any particular software libraries, packages, or their version numbers that were used for implementation. |
| Experiment Setup | Yes | In this section, we investigate the finite-sample performance of our estimator. We set τ = 2 and generate n i.i.d. copies of (Li, Ti, Yi) as follows: Li Unif(0, 1), P(Ti > t|Li) = exp{ exp(0.25Li)t}, Yi N(exp(1 1.5Li (2 Ti 2)), 0.52), where we observe {Li, Ti 2, i = 1(Ti < 2), Yi}1 i n. We examine the performance of the IPW estimator ˆψ(θ) and its variance estimate when θ(t, l) 1/3, 1/2.5, 1/2, 1/1.5, 1.5, 2, 2.5, 3 by reporting biases, percent biases (%Bias), empirical standard errors (SEE), average estimated standard errors (SD) based on B = 200 multiplier bootstrap (van der Vaart & Wellner, 1996; Kosorok, 2008), and coverage probabilities (95% CP) of Wald type 95% confidence intervals using R = 1000 simulated data sets of size n = 200, 1000, 5000. |