- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper proposes Neural-PCI, a neural framework for estimating longitudinal causal effects using outcome and treatment proxies to address unmeasured confounding. It jointly estimates recursively coupled bridge functions through maximum moment restriction losses and constructs outcome-regression, weighting, and targeted estimators. The paper presents error analyses and synthetic and semi-synthetic experiments supporting its proposed approach.
Critical strengths
The paper tackles an important practical challenge: making longitudinal proximal causal estimation flexible and computationally implementable beyond restrictive parametric bridge models. Its shared recurrent architecture and joint estimation framework provide a concrete approach to fitting coupled bridge equations, with a dedicated stabilization strategy for treatment bridges.
Critical weaknesses
- It seems that the identification assumptions are not demonstrated to hold in the synthetic experimental setting. For Assumption A.6, the paper mentioned that “The condition holds automatically when is Markov and structural equations depend on U only through its current value”. I’m not sure if this is true in general. For example, at , A.6 requires \begin{equation} U_0\perp \bigl(Y^{\bar a},W_0,Z_0,W_1,Z_1\bigr)\mid U_1,A_0,X_0,X_1. \end{equation} Consequently, it requires . Consider this simple model: \begin{equation} U_0 \sim N(0,1), \quad U_1 = \frac{1}{2}U_0 + \epsilon, \quad W_0 = U_0 + \eta \end{equation} where are independent standard normal noises, independent of . Let treatment be independently randomized and be constants. However, in this simple case, A.6 fails:
Questions for authors
-
For Assumption A.5, the synthetic DGP would require, among other things, \begin{equation} Z_t\perp W_{t-1}\mid U_t,\bar A_{t-1},\bar X_t, \end{equation} since its outcome proxies are unaffected by treatment interventions. However, Appendix H contains the path \begin{equation} Z_t\leftarrow Z_{t-1}\leftarrow U_{t-1}\rightarrow W_{t-1}, \end{equation} which is not blocked by the stated conditioning set. Thus, the graphical structure does not guarantee this conditional independence. Could the authors explain why the required conditional independence holds under their specific structural equations?
-
Why does the equivalence in Appendix E.1 transfer double robustness to the targeted estimator?
-
Why does Neural P-IPW improve as the horizon increases? Its MAE decreases from 0.054 at to 0.022 at , despite the increasing difficulty of cumulative weighting. What would be an intuitive interpretation of this result?
-
What explains the abrupt deterioration of MSMM P-OR at ? Its MAE rises from 0.076 at to 2.276 at $T=20, while MSMM P-IPW changes from 0.303 to 0.390.
3 comments
Thank you for your thorough response. Your clarifications and the revisions to the manuscript have sufficiently addressed my previous concerns. I appreciate the considerable effort you have devoted to addressing these issues. While I recognize and appreciate these improvements, I believe the paper would still benefit from further refinement and polishing. With this in mind, I will respectfully maintain my original score.
Thank you for your reply. We are glad that our clarifications and revisions have addressed your concerns. In the revised version, we will integrate the corrected results into the main text, give complete proofs of the corrected arguments in the appendix, and revise the presentation throughout for clarity. If there are particular parts you would like us to refine, we would be grateful to know, and we will address them. Thank you again.
Thank you for a careful review. W1, W2 and Q1 are correct, and each is corrected below; for W3 we ran additional experiments it calls for. The identification formulas (Eqs. 2 and 4) and the Neural P-OR and P-IPW rows of Table 1 are unchanged.
W1 and Q1. Assumption A.6 and the open path
Both points are correct and share one cause: a past proxy such as W0 depends on a past confounder U0, so conditioning on the current Ut cannot separate them; the Markov property of U does not help. The claim that A.6 "holds automatically" was wrong, and A.6 will be removed.
We return to the history form of Ying et al. (2023). A.2, A.5, A.7 and A.8 will be stated with the full latent history Uˉt, and A.6 replaced by the forward condition Zˉt−1⊥(Ut,Xt)∣Aˉt−1,Uˉt−1,Xˉt−1, used only in the treatment-bridge induction, so identification follows Ying et al. (2023). In the synthetic DGP, the path Zt←Zt−1←Ut−1→Wt−1 is blocked once Uˉt is conditioned, and the forward condition holds by d-separation.
W2. Remark G.3
You are right: its control by two neighbouring discrepancies was unjustified, as your example shows. The revised conversion still gives bridge-function L2 error, as Theorem F.1 does. It splits each stage-(t+1) error into the part that stage t can reproduce exactly and the rest. Stage t copies the reproducible part without raising dkt, so it is passed back. The rest cannot be copied; it shows up in dkt, so its effect is charged to stage t, once. Your constant shift is reproducible: stage 2, whose target is Y, sees it through dk2∝∣c∣, and it passes back unchanged, so the bound for h1 uses dk2 and holds.
For the comparison: sequential estimation fits each stage after the next is frozen, so it passes the whole error back, enlarged by up to κ per stage (Theorem F.1). A fit with small joint loss keeps every dkt small, so the rest stays small at every stage and only the reproducible part is passed back. When reproducible errors are not enlarged in total as they are passed back and the constants of each stage are bounded, the joint bridge error grows at most linearly in T. Details: our response to ojfN (W1).
W3. Separating ill-posedness from error propagation
We claim one advance, not two: a single ill-posed moment condition is solved by MMR (Mastouri et al., 2021; Kompa et al., 2022), which we take as given. Our contribution is estimating the chain jointly. To address your concerns on the fairness of the comparison in figure 2(a), we ran additional analysis as follows:
Same selection criterion. Both arms now select the regularization weight from the same grid by the validation MMR loss. Updated MAE of sequential NMMR at T=3,5,7,10,20: 0.013, 0.042, 0.261, 0.276, 0.598. Against the unchanged joint arm (Neural P-OR in Table 1), the arms agree at T=3,5, but from T=7 the sequential error is 5–8 times larger and still rising, so tuning does not explain the gap.
Where the error arises. Our simulator gives the true ATE but not the true bridges, so we check each stage's moment condition instead: on target-regime trajectories, the mean stage residual H^t+1−H^t should be zero. In the sequential arm these means have the same sign at almost every stage and add up. In the joint arm they sit in one to three stages and partly cancel.
So the gap comes from error accumulating along the chain, not from a single inversion.
Q2. Does E.1 transfer double robustness?
The targeted output of Eq. (18) contained the observed-treatment indicator and Q^t, so it was not a function of the outcome-side history, as Appendix E requires. The revised estimator is Eq. (7) with stage-wise self-normalized weights, obtained by shifting each stage's outcome bridge by a constant, and it is doubly robust; the formula, derivation and results are in our response to ojfN (W2).
Q3. Neural P-IPW improving with T
We would not read it as an improvement. The Monte Carlo SE of the ground truth is 0.018–0.026 (to be added to Table 1) and the P-IPW errors span only 0.022–0.055, so the fair reading is "does not degrade". A likely reason: the RMS normalization (Eq. 14) and the Hájek form (Eq. 17) use only the relative structure of the weights, so P-IPW is insensitive to the scale drift that grows with T.
Q4. MSMM P-OR at T=20
This was an error in our implementation of the baseline: its future-step enumeration was capped at 12 steps, a limit we kept from the reference code, which affects only the two MSMM cells at T=20. Without the cap, MSMM P-OR is 0.396 and MSMM P-DR is 0.399; results at T≤10 are unchanged. Table 1 and Section 5.2 will be corrected. Neural-PCI still has the best number in every column.
We hope these corrections address your concerns, and we would be glad to answer any further question.