Pith. sign in

REVIEW 4 major objections 4 minor 52 references

Covariate-Adjusted Deep Causal Learning for Heterogeneous Panel Data Models

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read CoDEAL, a covariate-adjusted deep causal learning procedure, imputes missing control outcomes in staggered-adoption panels and estimates unit-specific treatment effects, with an error bound that vanishes as the panel grows.

desk verdict A useful nonlinear factor panel method with strong simulations, but the central theorem is only stated, not proved in the available text, and the algorithm-theory link has a treatment-contamination gap that needs closing. read the letter →

arxiv 2505.20536 v1 pith:KW5627IN submitted 2025-05-26 stat.ML cs.LGecon.EMstat.ME

Guanhao Zhou , Yuefeng Han , Xiufan Yu This is my paper · ORCID
classification stat.MLcs.LGecon.EMstat.ME MSC 62G0862H2562G05
keywords causalpaneldatamodelscounterfactualestimationheterogeneoustreatmenteffectsmatrixcompletionmissingnotatrandommulti-outputautoencodersnonlinearfactordeepneuralnetworks
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a deep-learning estimator, CoDEAL, can impute the unobserved control outcomes in a staggered-adoption panel and thereby recover unit-specific average treatment effects on the treated. The payoff of the claim is practical: policies rolled out in different units at different times, such as vaccine mandates across U.S. states, can be evaluated without assuming linear factor structures, parallel trends, or missing-at-random outcomes. The paper models each observed outcome as a nonlinear covariate function plus a nonlinear latent factor term plus a unit-specific treatment effect, estimates the covariate function with a deep ReLU network and the factor structure with a multi-output autoencoder, and imputes missing counterfactuals by matrix completion. Theorem 2.1 states that the average squared imputation error is bounded by terms decaying in the time length, the number of units, and the covariate smoothness, up to logarithmic factors, which makes the unit-specific ATT estimator consistent.

What carries the argument

The load-bearing object is the multi-output autoencoder, a function class $G^{K_1}_{AE}$ made of one shared encoder $\rho: \mathbb{R}^N \to \mathbb{R}^{K_1}$ and $N$ independent decoders $\phi_i: \mathbb{R}^{K_1} \to \mathbb{R}$, fitted on the untreated entries of the covariate-adjusted matrix. It carries the argument by representing the nonlinear factor model $\phi_i^*(F_t^*)$ as a low-dimensional reconstruction problem: the encoder maps each observed column to the latent factor space, and the unit-specific decoders absorb cross-unit heterogeneity that a single-output autoencoder cannot. The second key mechanism is Assumption 3, a pervasiveness condition requiring a sparse matrix $M^*$ and a Hölder reconstruction map $\rho^*$ that invert the loading map on the factor range $[-B,B]^K$ up to order $n^{-1}$, which is what lets the autoencoder recover $F_t^*$. The covariate-adjustment step, equation (4), estimates $g_t^*$ on untreated entries before factor recovery so that covariate effects do not contaminate the latent structure.

What would settle it

Construct a four-block panel under the paper's model with $K=3$ latent factors, but let one factor load only on a small subset of units so Assumption 3 fails while the rest of the design is unchanged; run CoDEAL and record the average squared imputation error as $N$ and $T$ grow. If the error still follows the Theorem 2.1 rate, pervasiveness is not needed in the stated form; if the error stops shrinking or degrades sharply, the assumption is doing real work.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a panel outcome can be decomposed as $Y_{it} = \phi_i^*(F_t^*) + g_t^*(X_i) + \tau_i^*\,\mathbf{1}\{W_{it}=1\} + \epsilon_{it}$, where $g_t^*$ is a $(\beta,C)$-Hölder smooth function of covariates, $\phi_i^*$ is a unit-specific nonlinear loading on a $K$-dimensional latent factor $F_t^*$, and $\tau_i^*$ is the unit's ATT. CoDEAL removes $g_t^*$ with a fully connected ReLU network fitted on untreated entries, then fits a multi-output autoencoder -- one shared encoder and $N$ separate decoders -- on the covariate-adjusted control outcomes to recover the factors and loadings. The imputed control outcome is $\hat Y_{it}(0) = \hat\phi_i(\hat\rho(\tilde Y_{\cdot t})) + \hat g_t(X_i)$. Under the paper's pervasiveness assumption, the encoder can approximately invert the loading map, and Theorem 2.1 bounds the average squared distance between the imputed values and the true noiseless structure by $O_P\big((T^{-2\beta/(2\beta+K)} + T^{-1}n + N^{-1}K_1 + n^{-1})\log^4(NT) + N^{-2\beta/(2\beta+k)}\log^5 N\big)$. Consequently the plug-in estimator $\hat\tau_i = \sum_{t:\,W_{it}=1}\big(Y_{it}(1)-\hat Y_{it}(0)\big)/\sum_t W_{it}$ is consistent for $\tau_i^*$ in the four-block and general staggered-adoption designs.

Load-bearing premise

The load-bearing premise is the pervasiveness condition in Assumption 3: a sparse linear combination of the unit loadings, with coefficients of order $1/n$ spread over about $n$ units, can be inverted by a smooth map to recover the latent factors to within order $1/n$; if some factors load only on a few units, the autoencoder cannot recover them and the Theorem 2.1 rate collapses.

Editorial extensions

If this is right

  • Unit-specific ATTs are consistently estimated in staggered-adoption designs by reducing each treated block to a four-block subproblem and imputing its control outcomes, without requiring missing-at-random outcomes or a linear factor model.
  • In the four-block and staggered simulations, CoDEAL matches the best linear baseline in linear settings and reduces mean absolute error and mean squared error by roughly 10–40% when factor or covariate effects are nonlinear.
  • The nonlinear covariate and nonlinear factor components nest existing linear factor-model causal matrix completion methods as special cases, and the paper argues the same machinery extends to tensor or higher-order panel data.
  • In the real-data application, the estimated counterfactual says that without mandatory vaccination, confirmed cases in the policy states would have been about 12.2% higher and deaths about 34.6% higher by 11 December 2021.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: Theorem 2.1's rate contains $T^{-1}n$ and $N^{-2\beta/(2\beta+k)}$ terms, so practical convergence is governed by how well Assumption 3's reconstruction map recovers the factors; a validation-time estimate of $\sup_{x\in[-B,B]^K}\|\rho^*(M^*\phi^*(x))-x\|$ would tell users how much of the guarantee is actually available for a given panel.
  • Editorial inference: the method fits $N$ separate decoders, so its parameter count grows linearly with the number of units; a natural extension the paper does not pursue is to share decoder strength across units through a covariate-dependent or hierarchical parameterization for very large panels.
  • Editorial inference: because covariate functions are estimated only on untreated entries, a hold-out falsification exercise--reserve a pre-treatment block, impute it as missing, and compare with the observed entries--would isolate the covariate-adjustment and factor-recovery steps from treatment-effect estimation.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper proposes CoDEAL, a deep-learning method for estimating unit-specific average treatment effects on the treated (ATT) in staggered-adoption panel data. The model is Y_it = φ_i^*(F_t^*) + g_t^*(X_i) + τ_i^* W_it + ε_it, with a nonlinear factor structure φ_i^*(F_t^*) captured by a multi-output autoencoder and covariate effects g_t^*(X_i) captured by a deep ReLU network. The method first estimates covariate effects on untreated entries, forms covariate-adjusted outcomes, trains a multi-output autoencoder on untreated entries, and imputes missing counterfactuals by evaluating the trained autoencoder on the covariate-adjusted columns and adding back the covariate prediction. Theorem 2.1 claims an average squared error bound for imputed counterfactuals under Assumptions 1–3, and extensive simulations plus an OxCGRT COVID-19 policy application are reported.

Significance. If the theoretical guarantee and the empirical claims hold, CoDEAL would be a meaningful extension of factor-model-based causal matrix completion to nonlinear latent structures and unit-specific heterogeneity, with a practical contribution in covariate adjustment. The paper is clearly written and ships code with the supplementary material; the simulation study is extensive in breadth (four configurations, six covariate DGPs, four factor DGPs, 50 replications). The main weaknesses are that the central theorem is not verifiable from the submitted manuscript because the proof appendix is absent, and that the algorithm as written evaluates the learned encoder on inputs containing treated observations with O(1) treatment shifts, a setting the theorem does not cover.

major comments (4)
  1. [§2.3, Theorem 2.1] The proof and the required parameter choices are deferred to "Appendix B," but no appendix is included in the submitted manuscript. Since Theorem 2.1 is the central theoretical claim and its rate depends on unspecified DNN class parameters, the claim cannot be checked. A revision must include the full proof, the precise network architectures, and the explicit parameter settings that yield the stated rate.
  2. [§2.2.1, Algorithm 1 steps 4–5 and Eq. (6)] There is a mismatch between the trained model and the input used at prediction time. The autoencoder is fitted using only control entries in the loss (6), but step 5 evaluates the learned encoder on the full covariate-adjusted column \tilde{Y}_{·t}, which for treated periods contains entries of the form φ_i^*(F_t^*) + τ_i^* + ε_it for treated units. Theorem 2.1 bounds the imputation error with no term involving τ_i^*, so the rate cannot hold for the algorithm as stated unless one assumes the encoder is robust to O(1) treatment shifts on a subset of input coordinates, an assumption nowhere stated or implied by Assumptions 1–3. Please either modify the algorithm to de-contaminate the columns before encoding (for example by iterating imputation) or provide an explicit robustness argument with accompanying conditions.
  3. [§2.3, Assumption 3 and Theorem 2.1] Assumption 3 introduces an auxiliary matrix M^* and reconstruction map ρ^* that the algorithm never estimates or uses, and the theorem's rate contains an integer n whose relationship to N and T is not specified beyond "some diverging integer n > 0." Because the DNN encoder is trained by minimizing (6), it is not established that the learned encoder approximates ρ^* in the sense needed to convert Assumption 3 into the stated error bound. Please define n explicitly, state its relation to N and T, and show how the trained network class can realize the reconstruction within the error tolerance.
  4. [§3.4, Table 4 and accompanying text] The text says CoDEAL "consistently achieves the lowest MAE and MSE across nearly all settings," but in Table 4, Config. 3 with linear factors and r=5 or r=10, MC-NNM reports lower MAE (0.599 vs 0.614 and 0.616 vs 0.625, respectively) and lower MSE (0.685 vs 0.721 and 0.733 vs 0.754). The more cautious wording in the later paragraph, "performance comparable to MC-NNM in linear settings," is accurate; the earlier claim should be revised to match the table.
minor comments (4)
  1. [§3.3, Table 2] Several entries have unmatched parentheses, e.g., "1.517(0.030" in the log/linear-factor NN column and "1.216(0.013 2.790" in the poly/Vert-Reg row; these should be fixed.
  2. [§3.4, Table 4] In the Config. 2, nonlinear-covariate, polynomial row, the MC-NNM MSE is reported as "1.654(0.07)" with one decimal in the standard error, inconsistent with the other entries that use three decimals; please standardize.
  3. [§2.2.2, Eq. (7)] The definition of D(ξ0,η0) uses the index j for the time dimension while the surrounding text uses η; please replace j with η for consistency.
  4. [§4, Real data application] The policy effect estimates are presented without uncertainty quantification; a brief discussion of sensitivity or a bootstrap interval would help readers assess the stability of the counterfactual estimates.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the counterfactual imputation is fit only on untreated entries and Theorem 2.1's rate is not a restatement of any fitted quantity.

full rationale

The derivation chain is self-contained relative to its stated assumptions. The covariate functions are estimated by (4) using only untreated entries W_it=0, and the multi-output autoencoder is estimated by (6) using the same untreated entries; treated-period outcomes enter only when the fitted encoder/decoder is evaluated at prediction time in Algorithm 1 step 5 and in the ATT formula. The theorem's error bound is not a restatement of an objective the method optimizes: it mixes smoothness parameters, pervasiveness, sample sizes, and an approximation tolerance. The n^{-1} term in the rate is exactly the tolerance permitted in Assumption 3, but that is an identifiability/approximation assumption, not a parameter fitted from data and then renamed as a prediction. Self-citations (Han et al. 2024; Yu et al. 2024; Zhou et al. 2025; Chen et al. 2024, 2025) appear in related-work discussion and future extensions, and they do not carry Theorem 2.1 or the algorithm. The skeptic's concern about treatment contamination in the autoencoder input columns at treated periods identifies a potential gap between the theorem and Algorithm 1, since a complete proof would need an invariance or robustness argument for the treatment shifts present in the encoder input; however, that is a correctness gap rather than a circular reduction, because no equation in the paper forces the claimed bound to equal the training objective or a fitted parameter. The paper is also explicit in Section 5 about its binary-treatment limitation, which is a scope restriction, not circular reasoning.

Assumptions & free parameters 2 free parameters · 7 assumptions · 0 invented entities

No new physical or mechanistic entities are introduced; latent factors are inherited from the factor model literature. The main model ingredients are standard, and the central burden is carried by the strong functional form and identifiability assumptions in Assumptions 1 to 3.

free parameters (2)
  • Latent factor dimension K (and AE bottleneck K1) = K set to true values 3 or 4 in simulations; K not reported for real data
    Algorithm 2 requires K as input; the theorem assumes K1 >= K. No data-driven selection or sensitivity analysis is provided, so the method depends on a user-specified rank.
  • DNN architecture and training hyperparameters = not given
    The covariate DNN and multi-output AE require depth, widths, learning rate, epochs, regularization, and initialization. The theorem only cites a 'suitable choice of parameters' in Appendix B, and the main text omits these details.
assumptions (7)
  • domain assumption SUTVA and non-dynamic static treatment effects.
    Section 2.1 Causal Assumptions (i) and (ii); treatment at one unit or time does not affect other potential outcomes and effects do not evolve over time.
  • domain assumption Outcome model (1) is correctly specified: additive separable nonlinear factor component phi_i(F_t), covariate function g_t(X_i), and constant treatment effect tau_i.
    Equation (1) in Section 2.2 is assumed as the true data generating process; no misspecification testing is offered.
  • domain assumption Assumption 1: compact covariate support, bounded zero-mean factors, sub-Gaussian errors, mutual independence of epsilon, F_t, and X_i.
    Section 2.3 Assumption 1; used for DNN approximation and concentration arguments.
  • domain assumption Assumption 2: Holder smoothness of covariate functions g_t and later of rho_star.
    Section 2.3 Assumption 2; standard nonparametric smoothness but restricts the covariate effect class.
  • ad hoc to paper Assumption 3: pervasiveness and existence of sparse M_star and Holder rho_star with rho_star(M_star phi_star(x)) approximately x.
    Section 2.3 condition (8); tailored identifiability condition enabling factor recovery by the autoencoder. Hard to verify in applications.
  • domain assumption At least one never-treated unit and irreversible staggered adoption.
    Section 2.1 and Appendix A; needed to provide a control block for training and for the four-block construction.
  • domain assumption Treatment assignment and missingness are compatible with model (1): treated counterfactuals obey the same factor structure as controls.
    The imputation method assumes Y_it(0) follows (1) for all units, including treated units in treated periods. This is the parallel-trends substitute for causal interpretation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Covariate-Adjusted Deep Causal Learning for Heterogeneous Panel Data Models." pith.science (2026). https://pith.science/paper/KW5627IN

@misc{pith2026250520536,
  author       = {Pith},
  title        = {Pith review of: Covariate-Adjusted Deep Causal Learning for Heterogeneous Panel Data Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KW5627IN}},
  note         = {Machine review of arXiv:2505.20536}
}
read the original abstract

This paper studies the task of estimating heterogeneous treatment effects in causal panel data models, in the presence of covariate effects. We propose a novel Covariate-Adjusted Deep Causal Learning (CoDEAL) for panel data models, that employs flexible model structures and powerful neural network architectures to cohesively deal with the underlying heterogeneity and nonlinearity of both panel units and covariate effects. The proposed CoDEAL integrates nonlinear covariate effect components (parameterized by a feed-forward neural network) with nonlinear factor structures (modeled by a multi-output autoencoder) to form a heterogeneous causal panel model. The nonlinear covariate component offers a flexible framework for capturing the complex influences of covariates on outcomes. The nonlinear factor analysis enables CoDEAL to effectively capture both cross-sectional and temporal dependencies inherent in the data panel. This latent structural information is subsequently integrated into a customized matrix completion algorithm, thereby facilitating more accurate imputation of missing counterfactual outcomes. Moreover, the use of a multi-output autoencoder explicitly accounts for heterogeneity across units and enhances the model interpretability of the latent factors. We establish theoretical guarantees on the convergence of the estimated counterfactuals, and demonstrate the compelling performance of the proposed method using extensive simulation studies and a real data application.

Figures

Figures reproduced from arXiv: 2505.20536 by the authors.

Figure 1
Figure 1. A graphical illustration of the proposed CoDEAL in a four-block design. Gray and [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. Examples of constructing a four-block submatrix [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. Indicator matrix of the implementation of mandatory vaccination policy. [PITH_FULL_IMAGE:figures/full_fig_p023_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Comparison of the total confirmed cases (Left) and deaths (Right) across policy [PITH_FULL_IMAGE:figures/full_fig_p024_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

52 extracted references · 41 canonical work pages

  1. [1]

    Using synthetic controls: Feasibility, data requirements, and methodological aspects

    Alberto Abadie. Using synthetic controls: Feasibility, data requirements, and methodological aspects. Journal of Economic Literature, 59 0 (2): 0 391--425, 2021

    2021

  2. [2]

    The economic costs of conflict: A case study of the basque country

    Alberto Abadie and Javier Gardeazabal. The economic costs of conflict: A case study of the basque country. American Economic Review, 93 0 (1): 0 113--132, 2003

    2003

  3. [3]

    Synthetic control methods for comparative case studies: Estimating the effect of C alifornia's tobacco control program

    Alberto Abadie, Alexis Diamond, and Jens Hainmueller. Synthetic control methods for comparative case studies: Estimating the effect of C alifornia's tobacco control program. Journal of the American Statistical Association, 105 0 (490): 0 493--505, 2010

    work page 2010

  4. [4]

    Comparative politics and the synthetic control method

    Alberto Abadie, Alexis Diamond, and Jens Hainmueller. Comparative politics and the synthetic control method. American Journal of Political Science, 59 0 (2): 0 495--510, 2015

    work page 2015

  5. [5]

    Synthetic interventions

    Anish Agarwal, Devavrat Shah, and Dennis Shen. Synthetic interventions. arXiv preprint arXiv:2006.07691, 2020

    arXiv 2006

  6. [6]

    Causal matrix completion

    Anish Agarwal, Munther Dahleh, Devavrat Shah, and Dennis Shen. Causal matrix completion. In The Thirty Sixth Annual Conference on Learning Theory, pages 3821--3826. PMLR, 2023

    work page 2023

  7. [7]

    Causal models for longitudinal and panel data: A survey

    Dmitry Arkhangelsky and Guido Imbens. Causal models for longitudinal and panel data: A survey. The Econometrics Journal, 27 0 (3): 0 C1--C61, 2024

    work page 2024

  8. [8]

    Identification of average treatment effects in nonparametric panel models

    Susan Athey and Guido Imbens. Identification of average treatment effects in nonparametric panel models. arXiv preprint arXiv:2503.19873, 2025

    arXiv 2025

Show all 52 references
  1. [9]

    Design-based analysis in difference-in-differences settings with staggered adoption

    Susan Athey and Guido W Imbens. Design-based analysis in difference-in-differences settings with staggered adoption. Journal of Econometrics, 226 0 (1): 0 62--79, 2022

    2022

  2. [10]

    Matrix completion methods for causal panel data models

    Susan Athey, Mohsen Bayati, Nikolay Doudchenko, Guido Imbens, and Khashayar Khosravi. Matrix completion methods for causal panel data models. Journal of the American Statistical Association, 116 0 (536): 0 1716--1730, 2021

    2021

  3. [11]

    Inferential theory for factor models of large dimensions

    Jushan Bai. Inferential theory for factor models of large dimensions. Econometrica, 71 0 (1): 0 135--171, 2003

    2003

  4. [12]

    Panel data models with interactive fixed effects

    Jushan Bai. Panel data models with interactive fixed effects. Econometrica, 77 0 (4): 0 1229--1279, 2009

    2009

  5. [13]

    Matrix completion, counterfactuals, and factor analysis of missing data

    Jushan Bai and Serena Ng. Matrix completion, counterfactuals, and factor analysis of missing data. Journal of the American Statistical Association, 116 0 (536): 0 1746--1763, 2021

    2021

  6. [14]

    Estimation and inference for cp tensor factor models

    Bin Chen, Yuefeng Han, and Qiyang Yu. Estimation and inference for cp tensor factor models. arXiv preprint arXiv:2406.17278, 2024

    2024

  7. [15]

    Diffusion index forecast with tensor data

    Bin Chen, Yuefeng Han, and Qiyang Yu. Diffusion index forecast with tensor data. Available at SSRN 5202316, 2025

    2025

  8. [16]

    Matrix completion when missing is not at random and its applications in causal panel data models

    Jungjun Choi and Ming Yuan. Matrix completion when missing is not at random and its applications in causal panel data models. Journal of the American Statistical Association, pages 1--15, 2024

    2024

  9. [17]

    On synthetic difference-in-differences and related estimation methods in stata

    Damian Clarke, Daniel Paila \ n ir, Susan Athey, and Guido Imbens. On synthetic difference-in-differences and related estimation methods in stata. The Stata Journal, 24 0 (4): 0 557--598, 2024

    2024

  10. [18]

    Balancing, regression, difference-in-differences and synthetic control methods: A synthesis

    Nikolay Doudchenko and Guido W Imbens. Balancing, regression, difference-in-differences and synthetic control methods: A synthesis. Technical report, National Bureau of Economic Research, 2016

    2016

  11. [19]

    Matrix completion by deep matrix factorization

    Jicong Fan and Jieyu Cheng. Matrix completion by deep matrix factorization. Neural Networks, 98: 0 34--41, 2018

    2018

  12. [20]

    Neuron-enhanced autoencoder matrix completion and collaborative filtering: Theory and practice

    Jicong Fan, Rui Chen, Zhao Zhang, and Chris Ding. Neuron-enhanced autoencoder matrix completion and collaborative filtering: Theory and practice. In The Twelfth International Conference on Learning Representations, 2024

    2024

  13. [21]

    A global panel database of pandemic policies (oxford covid-19 government response tracker)

    Thomas Hale, Noam Angrist, Rafael Goldszmidt, Beatriz Kira, Anna Petherick, Toby Phillips, Samuel Webster, Emily Cameron-Blake, Laura Hallas, Saptarshi Majumdar, and Helen Tatlow. A global panel database of pandemic policies (oxford covid-19 government response tracker). Natur...

    2021

  14. [22]

    Tensor principal component analysis in high dimensional cp models

    Yuefeng Han and Cun-Hui Zhang. Tensor principal component analysis in high dimensional cp models. IEEE Transactions on Information Theory, 69 0 (2): 0 1147--1167, 2022

    2022

  15. [23]

    Tensor factor model estimation by iterative projection

    Yuefeng Han, Rong Chen, Dan Yang, and Cun-Hui Zhang. Tensor factor model estimation by iterative projection. The Annals of Statistics, 52 0 (6): 0 2641--2667, 2024 a

    2024

  16. [24]

    Cp factor model for dynamic tensors

    Yuefeng Han, Dan Yang, Cun-Hui Zhang, and Rong Chen. Cp factor model for dynamic tensors. Journal of the Royal Statistical Society Series B: Statistical Methodology, 86 0 (5): 0 1383--1413, 2024 b

    2024

  17. [25]

    Estimating causal effects from panel data with dynamic multivariate panel models

    Jouni Helske and Santtu Tikka. Estimating causal effects from panel data with dynamic multivariate panel models. Advances in Life Course Research, 60: 0 100617, 2024

    2024

  18. [26]

    Impact of vaccination on infection or death from covid-19 in individuals with laboratory-confirmed cases: Case-control study

    Porfirio Felipe Hern \'a ndez Bautista, Concepci \'o n Grajales Mu \ n iz, David Alejandro Cabrera Gayt \'a n, Teresita Rojas Mendoza, Alfonso Vallejos Par \'a s, Clara Esperanza Santacruz Tinoco, Julio Elias Alvarado Yaah, Yu Mei Anguiano Hern \'a ndez, Nancy Sandoval Guti \'...

    2023

  19. [27]

    On the use of two-way fixed effects regression models for causal inference with panel data

    Kosuke Imai and In Song Kim. On the use of two-way fixed effects regression models for causal inference with panel data. Political Analysis, 29 0 (3): 0 405--415, 2021

    2021

  20. [28]

    Causal inference in the social sciences

    Guido W Imbens. Causal inference in the social sciences. Annual Review of Statistics and Its Application, 11: 0 123--152, 2024

    2024

  21. [29]

    Causal inference in statistics, social, and biomedical sciences

    Guido W Imbens and Donald B Rubin. Causal inference in statistics, social, and biomedical sciences. Cambridge University Press, 2015

    2015

  22. [30]

    Inverse moment methods for sufficient forecasting using high-dimensional predictors

    Wei Luo, Lingzhou Xue, Jiawei Yao, and Xiufan Yu. Inverse moment methods for sufficient forecasting using high-dimensional predictors. Biometrika, 109 0 (2): 0 473--487, 2022

    2022

  23. [31]

    Estimated number of lives directly saved by covid-19 vaccination programmes in the who european region from december, 2020, to march, 2023: a retrospective surveillance study

    Margaux M I Meslé, Jeremy Brown, Piers Mook, Mark A Katz, José Hagan, Roberta Pastore, Bernhard Benka, Monika Redlberger-Fritz, Veerle Bossuyt, Nathalie Stouten, et al. Estimated number of lives directly saved by covid-19 vaccination programmes in the who european region from ...

    2020

  24. [32]

    Fixed effects and causal inference

    Daniel L Millimet and Marc Bellemare. Fixed effects and causal inference. Technical report, IZA Discussion Papers, 2023

    2023

  25. [33]

    Generalized causal tree for uplift modeling

    Preetam Nandy, Xiufan Yu, Wanjun Liu, Ye Tu, Kinjal Basu, and Shaunak Chatterjee. Generalized causal tree for uplift modeling. In 2023 IEEE International Conference On Big Data (BigData), pages 788--798. IEEE, 2023

    2023

  26. [34]

    Simple, fast, and flexible framework for matrix completion with infinite width neural networks

    Adityanarayanan Radhakrishnan, George Stefanakis, Mikhail Belkin, and Caroline Uhler. Simple, fast, and flexible framework for matrix completion with infinite width neural networks. Proceedings of the National Academy of Sciences, 119 0 (16): 0 e2115064119, 2022

    2022

  27. [35]

    The central role of the propensity score in observational studies for causal effects

    Paul R Rosenbaum and Donald B Rubin. The central role of the propensity score in observational studies for causal effects. Biometrika, 70 0 (1): 0 41--55, 1983

    1983

  28. [36]

    Inference on heterogeneous treatment effects in high-dimensional dynamic panels under weak dependence

    Vira Semenova, Matt Goldman, Victor Chernozhukov, and Matt Taddy. Inference on heterogeneous treatment effects in high-dimensional dynamic panels under weak dependence. Quantitative Economics, 14 0 (2): 0 471--510, 2023

    2023

  29. [37]

    Causal imputation via synthetic interventions

    Chandler Squires, Dennis Shen, Anish Agarwal, Devavrat Shah, and Caroline Uhler. Causal imputation via synthetic interventions. In Conference on Causal Learning and Reasoning, pages 688--711. PMLR, 2022

    2022

  30. [38]

    Estimating dynamic treatment effects in event studies with heterogeneous treatment effects

    Liyang Sun and Sarah Abraham. Estimating dynamic treatment effects in event studies with heterogeneous treatment effects. Journal of Econometrics, 225 0 (2): 0 175--199, 2021

    2021

  31. [39]

    Introduction \`a l'estimation non param \'e trique , volume 41

    Alexandre B Tsybakov. Introduction \`a l'estimation non param \'e trique , volume 41. Springer Science & Business Media, 2003

    2003

  32. [40]

    Estimation and inference of heterogeneous treatment effects using random forests

    Stefan Wager and Susan Athey. Estimation and inference of heterogeneous treatment effects using random forests. Journal of the American Statistical Association, 113 0 (523): 0 1228--1242, 2018

    2018

  33. [41]

    Designing difference-in-difference studies with staggered treatment adoption: Key concepts and practical guidelines

    Coady Wing, Madeline Yozwiak, Alex Hollingsworth, Seth Freedman, and Kosali Simon. Designing difference-in-difference studies with staggered treatment adoption: Key concepts and practical guidelines. Annual Review of Public Health, 45: 0 485–505, 2024

    2024

  34. [42]

    Nana Wu, Keven Joyal-Desmarais, Paula AB Ribeiro, Ariany Marques Vieira, Jovana Stojanovic, Comfort Sanuade, Doro Yip, and Simon L Bacon. Long-term effectiveness of covid-19 vaccines against infections, hospitalisations, and mortality in adults: findings from a rapid living sy...

    2022

  35. [43]

    Large dimensional latent factor modeling with missing observations and applications to causal inference

    Ruoxuan Xiong and Markus Pelger. Large dimensional latent factor modeling with missing observations and applications to causal inference. Journal of Econometrics, 233 0 (1): 0 271--301, 2023

    2023

  36. [44]

    Deep autoencoders for nonlinear factor models: Theory and applications

    Dacheng Xiu and Zhouyu Shen. Deep autoencoders for nonlinear factor models: Theory and applications. Available at SSRN, 2024

    2024

  37. [45]

    A systematic review and meta-analysis of the effectiveness and safety of covid-19 vaccination in older adults

    Kun Xu, Zihan Wang, Maorong Qin, Yangyu Gao, Na Luo, Wanting Xie, Yihan Zou, Jie Wang, and Xingming Ma. A systematic review and meta-analysis of the effectiveness and safety of covid-19 vaccination in older adults. Frontiers in Immunology, 14: 0 1113156, 2023

    2023

  38. [46]

    Generalized synthetic control method: Causal inference with interactive fixed effects models

    Yiqing Xu. Generalized synthetic control method: Causal inference with interactive fixed effects models. Political Analysis, 25 0 (1): 0 57--76, 2017

    2017

  39. [47]

    Nonlinear factor analysis as a statistical method

    Ilker Yalcin and Yasuo Amemiya. Nonlinear factor analysis as a statistical method. Statistical Science, 16 0 (3): 0 275--294, 2001

    2001

  40. [48]

    Entrywise inference for missing panel data: A simple and instance-optimal approach

    Yuling Yan and Martin J Wainwright. Entrywise inference for missing panel data: A simple and instance-optimal approach. arXiv preprint arXiv:2401.13665, 2024

    2024 arXiv

  41. [49]

    Dynamic matrix factor models for high dimensional time series

    Ruofan Yu, Rong Chen, Han Xiao, and Yuefeng Han. Dynamic matrix factor models for high dimensional time series. arXiv preprint arXiv:2407.05624, 2024 a

    2024 arXiv

  42. [50]

    Nonparametric estimation and conformal inference of the sufficient forecasting with a diverging number of factors

    Xiufan Yu, Jiawei Yao, and Lingzhou Xue. Nonparametric estimation and conformal inference of the sufficient forecasting with a diverging number of factors. Journal of Business & Economic Statistics, 40 0 (1): 0 342--354, 2022

    2022

  43. [51]

    Power enhancement for testing multi-factor asset pricing models via fisher’s method

    Xiufan Yu, Jiawei Yao, and Lingzhou Xue. Power enhancement for testing multi-factor asset pricing models via fisher’s method. Journal of Econometrics, 239 0 (2): 0 105458, 2024 b

    2024

  44. [52]

    Factor augmented tensor-on-tensor neural networks

    Guanhao Zhou, Yuefeng Han, and Xiufan Yu. Factor augmented tensor-on-tensor neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 22928--22936, 2025

    2025

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.