- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
The paper proposed a way of leveraging the proximal causal learning to solve the problem of unobserved confounders. Providing two identification results with two different proxies.
Critical strengths
- The paper is well-written and there is paras before different section serves as the connections which help the readers understand.
- The experiments compared with a lot of the baselines.
Critical weaknesses
- In the experiment section, there is only simulation studies and synthetic experiments. Which is not enough, should at least has semi-synthetic experiments or real-world experiments to show that the results of the methods.
- Missing citations about the decision making for example: for the potential outcome framework, did you follow the Rubin 1974's framework or any other streams. Also for example, in the assumption could cite some well-known papers instead just saying that "general assumptions in the causal inference literature".
- Before identifying your target estimand, the population where the data is sampled from is missing.
- Minor: There is a space missing in Line 322, "our proposed estimators.We".
Questions for authors
- To leverage the proxy causal framewok, one of the preliminary is that you could dividing L into three parts, how is step is conducted? Does the author pre-decide the three subsets X, W, Z, or the algorithm learn it by it self. And how different division will affect the estimation results.
- Could the author shows the confidence intervals of the proposed methods, do the CIs could cover the oracles with valid coverage percent?
- I would like to ask about sensitivity analysis of the bandwidth of the kernel estimation. Since if the bandwidth is too small, there will be hard to get enough sample to estimate with large variance, and if bandwidth is too large then the bias will be a problem.
3 comments
Q1: How is dividing into conducted? Pre-decided or learned, and how does different division affect estimation?
We thank the reviewer for this important question. Following the standard proximal causal inference literature (Miao et al., 2018; Tchetgen Tchetgen et al., 2024; Cui et al., 2024), the partitioning of into is pre-specified using domain knowledge and causal/temporal ordering rather than learned purely data-driven, because the conditional independence conditions in Assumption 1 and ) are generally untestable from observational data alone. contains measured adjustment variables. For , we need a pre-treatment measurement informative about the latent confounder whose association with dose is accounted for by . For , we need a measurement informative about treatment assignment with no direct effect on the outcome after accounting for .
To illustrate how this is done in practice, consider our MIMIC-IV semi-synthetic study (Section 7) on initial norepinephrine dosing in ICU patients:
- Observed confounders : Lactate, blood urea nitrogen, bicarbonate, creatinine, potassium, and the missing-immature-granulocyte indicator capture acute metabolic acidosis and renal/perfusion status, which directly guide vasopressor dosing () and affect short-term blood pressure recovery ().
- Outcome-inducing proxy (pre-infusion platelet count): Platelet count reflects underlying sepsis severity and coagulopathy () and correlates with hemodynamic outcomes (). Because it is measured prior to infusion initiation, it cannot be affected by , and conditional on it does not directly drive the initial norepinephrine infusion rate—making clinically plausible.
- Treatment-inducing proxies (white blood cell count and neutrophil percentage): Measured at blood-culture sampling before norepinephrine initiation, these inflammatory markers reflect acute infection severity () and influence clinicians' assessment of septic shock and initial dosing (), while having no direct physiological effect on 1-hour mean arterial pressure () once are accounted for—making clinically plausible.
Regarding how different divisions affect estimation:
- Misclassifying variables such that Assumption 1 is violated can introduce bias (as discussed in Section 8).
- Within valid treatment-inducing proxies , swapping which variable enters versus serving as the auxiliary proxy affects finite-sample performance: as shown in Section 6 (Figure 2), assigning the more informative proxy () to yields better performance than the reverse.
Q2: Could the author show the confidence intervals of the proposed methods, and do the CIs cover the oracles with valid coverage percent?
We thank the reviewer for raising this question on statistical inference. Our theoretical analysis in Section 5 establishes a finite-sample high-probability excess-risk (regret) bound for (Theorem 4) rather than an asymptotic normality theorem for or . Deriving asymptotic normality and valid confidence intervals for non-regular, nonsmooth continuous-dose O-learning estimators—which involve a vanishing kernel bandwidth and RKHS regularization—is a challenging theoretical problem beyond the scope of this work. Empirically, the boxplots over 100 replications (Figures 2, 3, 5, 7, 8) and the means standard deviations in Table 5 reflect the finite-sample dispersion of our estimators relative to .
Q3: Sensitivity analysis of the bandwidth of the kernel estimation (bias-variance trade-off).
We thank the reviewer for this insightful suggestion. We have added a dedicated sensitivity analysis in Appendix E.4 (Figure 6, pp. 29–30), evaluating across for under both Scenario 1 and Scenario 2 at . Consistent with the reviewer's intuition, Figure 6 exhibits a clear bias-variance trade-off: small (e.g., ) yields high variance due to sparse local samples in the kernel window, large (e.g., ) increases smoothing bias, and the policy value peaks at .
References:
- Cui, Y., Pu, H., Shi, X., Miao, W., & Tchetgen Tchetgen, E. (2024). Semiparametric proximal causal inference. Journal of the American Statistical Association, 119(546), 1348–1359.
- Hernán, M. A., & Robins, J. M. (2006). Estimating causal effects from epidemiological data. Journal of Epidemiology and Community Health, 60(7), 578–586.
- Kennedy, E. H., Ma, Z., McHugh, M. D., & Small, D. S. (2017). Non-parametric methods for doubly robust estimation of continuous treatment effects. Journal of the Royal Statistical Society Series B: Statistical Methodology, 79(4), 1229–1245.
- Miao, W., Geng, Z., & Tchetgen Tchetgen, E. J. (2018). Identifying causal effects with proxy variables of an unmeasured confounder. Biometrika, 105(4), 987–993.
- Rubin, D. B. (1974). Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of Educational Psychology, 66(5), 688.
- Tchetgen Tchetgen, E. J., Ying, A., Cui, Y., Shi, X., & Miao, W. (2024). An introduction to proximal causal inference. Statistical Science, 39(3), 375–390.
Thank you for the encouraging feedback on our writing and experimental comparisons, as well as the helpful suggestions. We address each point below.
W1: Only simulation studies and synthetic experiments; should at least have semi-synthetic or real-world experiments.
We thank the reviewer for this valuable suggestion. In the revised manuscript, we have added a semi-synthetic experiment based on real-world ICU records from MIMIC-IV v3.1 in Section 7 (with full DGP details in Appendix G and results in Figure 3), evaluating initial norepinephrine dosing for critically ill adults. All six proposed proximal rules outperform all competing baselines, achieving higher policy values while avoiding the upward confounding bias of Naive methods and the extreme boundary dosing of PMMR, VCNet-TR, CPO-DR, and CPO-IPW.
W2: Missing citations about decision making (e.g., Rubin 1974 for potential outcomes; well-known papers for general assumptions).
We thank the reviewer for catching these missing references. We have added the missing citations in Section 2 and Section 3.1, explicitly citing
Rubin (1974)for the potential-outcomes framework andHernán & Robins (2006); Kennedy et al. (2017)for consistency, positivity, and unconfoundedness.W3: Before identifying your target estimand, the population where the data is sampled from is missing.
We thank the reviewer for pointing out this omission. In Section 2, we now explicitly state: "The observed data {(Li,Ai,Yi)}i=1n are independent and identically distributed draws from the observed-data distribution of this population. Our objective is to learn a dose rule for new units drawn from the same population." (See also Section 3.)
W4: Minor: There is a space missing in Line 322, "our proposed estimators.We".
We thank the reviewer for spotting this typo. We have fixed this spacing typo in Section 5.