- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper studies confidence intervals for a population mean under differential privacy. It first clips the observations and analyzes the resulting bias, as well as the non-identifiability of the original mean under fixed clipping without additional assumptions. It then partitions the data into disjoint blocks, adds Laplace noise to the block means, and applies empirical likelihood and adjusted empirical likelihood to construct confidence intervals. The theoretical results establish asymptotic chi-squared distributions for the likelihood-ratio statistics at the clipped population mean. The paper also discusses an extension to the original mean using a bound on the clipping bias. The experiments compare several methods across different distributions, including heavy-tailed settings, and include real-data applications.
Critical strengths
- The paper explicitly distinguishes the clipped population mean from the original mean and analyzes the consequences of this distinction. The treatment of clipping bias and non-identifiability helps clarify what the proposed procedure can actually infer.
- The paper is relatively concise, and the empirical evaluation covers several competing methods, different distributional settings, and real-data applications. Reporting both coverage and interval length provides useful evidence about the finite-sample behavior of EL and AEL.
Critical weaknesses
- The additional methodological and theoretical insight is not sufficiently clear. The construction combines clipping, privatized block means, and established EL/AEL procedures. Given the relative simplicity of this construction, I would expect a clearer explanation of what we learn beyond the validity of this particular combination. The paper emphasizes a Wilks-type result for a triangular array with vanishing row variance, but it is not sufficiently clear which aspects require substantially new arguments beyond existing empirical-likelihood theory. This makes the significance of the theoretical contribution difficult to assess.
- The cost of privacy and the resulting efficiency trade-offs need a more explicit characterization. The paper discusses privacy-noise variance and varies the privacy budget in Table 2, but these results do not sufficiently clarify the loss relative to a comparable non-private procedure or separate the cost of privacy from the additional cost introduced by the block construction. In particular, the number of blocks affects both calibration and privacy noise. A clearer account of how sample size, privacy budget, block count, and clipping bounds jointly determine interval length and coverage would substantially strengthen the contribution.
Questions for authors
- The algorithm is relatively simple, but I do not see sufficient new insight beyond applying existing EL/AEL methods to privatized block means. The additional contribution needs to be clearer.
- My main concern is the cost of privacy. Although the paper varies the privacy budget experimentally, it does not sufficiently explain the trade-off between privacy and statistical efficiency or quantify the loss relative to a non-private baseline.
- Figure 1 is difficult to interpret because the curves are barely distinguishable. I cannot clearly see the differences between the settings or how much AEL improves over EL.
3 comments
Response to Reviewer 6L8e (Cont.)
Response to W2 / Q2
We separate the cost of blocking from the cost of privacy, first in closed form and then empirically.
With B equal blocks of size m = n/B that use all n observations, clipping width w and , the average of the privatized block means (the point at which the EL statistic is zero) has exact variance
- Blocking alone preserves the sample mean and its variance exactly. The sampling term does not depend on B; finite-sample EL/AEL calibration and interval length can still change with B.
- Privacy adds a variance inflation that grows linearly in B and in .
- Under the paper's asymptotics (, fixed and w), the privacy-noise variance is of order , i.e. its standard deviation is . So when , EL/AEL are first-order efficient relative to the non-private sampling variance. The decomposition concerns ; finite-sample interval length also reflects calibration, and the privacy term matters most at small .
Empirical decomposition (n = 400, Normal(1,1), 2,000 replications):
| Method | = 1 | = 2 | = 3 | = 4 | = 5 | Coverage | |
|---|---|---|---|---|---|---|---|
| Non-private, 400 observations | EL | 0.188 | 0.188 | 0.188 | 0.188 | 0.188 | 0.946 |
| AEL | 0.190 | 0.190 | 0.190 | 0.190 | 0.190 | 0.948 | |
| Non-private, 20 block means | EL | 0.186 | 0.186 | 0.186 | 0.186 | 0.186 | 0.941 |
| AEL | 0.204 | 0.204 | 0.204 | 0.204 | 0.204 | 0.958 | |
| Non-private, 80 block means | EL | 0.189 | 0.189 | 0.189 | 0.189 | 0.189 | 0.948 |
| AEL | 0.195 | 0.195 | 0.195 | 0.195 | 0.195 | 0.954 | |
| DP, B = 20 | EL | 0.31 | 0.22 | 0.20 | 0.20 | 0.19 | 0.925–0.945 |
| AEL | 0.34 | 0.24 | 0.22 | 0.21 | 0.21 | 0.949–0.963 | |
| DP, B = 80 | EL | 0.54 | 0.31 | 0.25 | 0.23 | 0.21 | 0.938–0.945 |
| AEL | 0.55 | 0.32 | 0.26 | 0.23 | 0.22 | 0.946–0.949 |
- Cost of blocking vs. cost of privacy. Comparing rows with the same method separates the two. Blocking alone costs essentially nothing for EL (0.186 and 0.189 against 0.188); for AEL it adds a little at B = 20 (0.204 against 0.190), the price of its small-sample adjustment. Privacy is the main cost, and it falls quickly with : at = 1 the DP intervals are about 1.6–2.9 times the non-private length, and at = 5 only 2–16% longer than the same method on all 400 clipped observations. Against non-private AEL on the same blocks, the = 5 privacy cost is 3% at B = 20 and 13% at B = 80. AEL covers at 0.946–0.963 in every DP row; EL covers at 0.925–0.945 and is below nominal at B = 20, = 1.
- The formula approximates these lengths. For B = 20 at = 1, the variance factor is . To first order this gives an EL length of , against 0.31 observed for EL and 0.34 for AEL.
- How the quantities interact:
- larger n and larger shrink the privacy term;
- larger B inflates it, but small B worsens calibration: at B = 20, plain EL covers at 0.925–0.945, while AEL covers at 0.949–0.963 (table above). AEL therefore allows a smaller B and a shorter interval: at B = 20 its lengths are 0.21–0.34, against 0.22–0.55 at B = 80;
- a wider clipping range inflates it through (and changes ), so the interval lengthens while coverage of stays at 0.947–0.955 (Table 4 of the paper), but it reduces the tail-clipping error bound of Prop. 2.1.
Response to Q3
- Figure 1 checks the approximation of Theorems 2.3 and 2.5 in finite samples. The curves overlap because the empirical distributions are close to the limit, as the theorems predict.
- The figure also shows that AEL is closer to than EL. The rejection rates at the nominal levels make this concrete :
| Nominal level | EL, n = 200/400/600 | AEL, n = 200/400/600 |
|---|---|---|
| 10% | 12.3 / 11.8 / 11.1 | 10.4 / 10.8 / 10.5 |
| 5% | 6.7 / 5.6 / 5.4 | 5.4 / 4.8 / 4.8 |
| 1% | 1.8 / 1.5 / 1.2 | 1.3 / 1.2 / 1.1 |
EL over-rejects, most at n = 200 (6.7% at the 5% level), while AEL is closer to the nominal level in all nine comparisons. This is the finite-sample improvement of AEL, and it matches the coverage gap between EL and AEL in Table 1.
Thank you for the detailed response. The variance decomposition and the non-private comparisons address my concern about the cost of privacy. The rejection rates clarify the improvement in AEL calibration that was difficult to see in Figure 1. I also appreciate the comparison with the noise-aware studentized interval and the explanation of the additional proof steps for privatized block means.
My remaining concern is the significance of the theoretical contribution. Although the response clarifies why this setting requires additional analysis, I still view the advance over existing EL/AEL theory as modest. I therefore maintain my original recommendation of weak rejection.
Response to Reviewer 6L8e
Response to W1 / Q1
What we learn beyond the validity of this combination.
The adjusted intervals achieve empirical coverage of 0.947–0.973 for the raw mean, compared with 0.946–0.959 for the unadjusted intervals. The principal benefit of the adjustment is that it explicitly accounts for clipping bias under an additional moment assumption, providing a theoretically justified extension of inference to the raw mean.
Which aspects require new arguments.
The theoretical contribution is the Wilks theorems for EL and AEL on privatized block means (Theorems 2.3 and 2.5). The released values x~j are block means plus Laplace noise, with variance σB2=σc2/m+2w2/(m2ϵ2)→0. They form a triangular array whose law changes with n and whose noise has heavier-than-Gaussian tails, so the classical fixed-distribution Wilks theorem does not apply directly, and the existing DP EL theory (Liu & Li, 2025) concerns local estimates released by separate machines. To establish the χ2 limits for this array, we prove: