- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes a new instrumented PCA (IPCA) approach based on graph convolutional networks (GCNs). From technical viewpoints, the method is extended from IPCA [Kelly+19], which models dynamics of temporal signals by linear factors, through non-linearly describing those factors by means of GCNs to effectively incorporate relational characteristics of micro-level instruments as well as macro-level public ones. The authors also provide theoretical analysis on the proposed model about the learning bound for the GCN-based model and asymptotic normality of its inference. In the experiments on the economical assets data, the proposed method exhibits competitive performance to the other asset-pricing models in terms of predictive performance and economic portfolio evaluation.
Critical strengths
- In the framework of IPCA [Kelly+19], GCNs are effectively introduced to non-linearly extend factor-related representations while exploiting two types of relational information embedded in micro-/macro-level instruments.
- The analysis in [Farrell+25] for fully-connected networks (FCNs) is extended to the proposed SGCN model for theoretical analyzing the asymptotic behaviors of the method. It theoretically supports the effectiveness of SGCN.
- The proposed method is evaluated on real-world stock price data to produce favorable economical performance outperforming the others. Thus, the theoretical and practical stuffs are well balanced in the paper.
Critical weaknesses
-
The theoretical aspects in the paper are heavily overlapped with [Farrell+25], thus leading to limited novelty. Specifically, Sec.4 basically follows the approach of [Farrell+25]; while the analysis regarding learning bound (Sec.4.1) includes extension to cope with the GCN models, the asymptotic normality in Sec.4.2 seems to be rather directly stem from the prior work.
-
In comparison to IPCA formulation (1), apart from GCN-based modeling, the proposed approach (7) additionally leverages macro-level public instrument to describe -term in IPCA, while the other comparison approaches, such as the ones reviewed in Sec.2, seem to be solely built upon the micro-level ones . Thus, the comparison experiments in Sec.5 contains concerns regarding fairness. To genuinely evaluate the efficacy of GCN-empowered modeling, the methods should be compared in the same framework using the identical input information. It also lacks clear comparison to simple baseline that replaces GCNs with feed-forward network models (MLPs); SDNN or self-only in Table A4 correspond to that?
-
The discussion about interpretability in Sec.4.3 is less convincing. While Theorem 2 provides effective bound of parameter learning in the GCN models, it is less helpful for interpreting the uniqueness of the model (parameters) in conjunction with the factors ; namely, one could conjecture that there could be another for rotated factor through joint learning of GCNs and factors in IPCA framework. Thus, the ambiguity in the factor model is not fully resolved in the approach. Furthermore, it is hard to find out that the normality shown in Theorem 3 contributes to interpretable factors. The interpretability of the factors seems to be less analyzed even in the experiments.
-
Overall optimization for this model is not clearly shown. As to architecture shown in Fig.1, it is unclear how the Bottleneck Factor Network operates in the proposed model (7). The factor network and its input portfolio return vector are not explained at all in the main texts, only depicted in Fig.1.
Questions for authors
- From theoretical aspects, please show clear novelty over the prior work [Farrell+25].
- Please provide clear analysis/discussion regarding the interpretability, hopefully with empirical (quantitative) results.
- Please discuss the fairness in comparison especially regarding the input data, such as .
- Please detail the overall framework such as for the factor network.
7 comments
Reply to Reviewer [2/3]
2b. A direct and fully matched MLP comparison
The self-only configuration is exactly the matched MLP control. the revised Appendix A.14 makes this equivalence explicit: setting the normalized adjacency to turns each graph layer into a pointwise linear map while retaining the direct characteristic path, readout, alpha branch, factor network, inputs, parameter count, and training protocol. Transferring the trained self-only weights to linear layers produces a maximum absolute difference of in alpha, beta, and return predictions across , tested on random batches and four monthly stock cross-sections.
The two models have identical parameter counts at every factor dimension, ranging from 11346 to 11591 over . Table A9 reports all six factor dimensions over the same 420 forecast months. Averaged over , FF49-based HeterIPCA raises stock-level from to , portfolio-level from to , net long-short Sharpe from to , and net long-only Sharpe from to . Both portfolio Sharpe ratios improve at every factor dimension. At , the long-short Sharpe improvement is , with a paired 95% stationary-bootstrap interval of based on 10000 replications and a mean block length of six months.
The graph alternatives in Table A4 provide a complementary structural comparison. Their mean portfolio-level values are for FF49, for two-digit SIC, for self-only, for a density-matched random graph, and for the characteristic-similarity graph. In summary, the matched MLP and alternative-graph results identify economically structured industry aggregation as a productive component of the model.
3. Interpretability: algebraic rotation is not the same as economic meaning
We respectfully clarify that Section 4.3 in the previous manuscript has already explicitly acknowledged the rotation cited by the reviewer. It writes
We therefore do not overlook rotational equivalence. Our point is different: once an economically grounded nonlinear instrument-to-loading mapping has been learned, an arbitrary post-hoc rotation does not create additional economic information. It changes the basis used to express the same systematic component and is unnecessary for explaining which observable characteristics are related to the learned loading representation. This is the sense in which Section 4.3 calls an additional arbitrary rotation redundant.
This distinction can be expressed formally. Let have full column rank, and let be invertible. The orthogonal projector onto the loading subspace is
Since ,
Thus subspace-based interpretation of the characteristic-to-loading relation is exactly invariant under the reviewer's transformation. A different coordinate system cannot nullify a characteristic association computed from this projection. The argument is about the meaningfulness of the learned representation, not a claim that nonlinear functions automatically prohibit rotated functions. Theorem 2 gives estimation accuracy conditional on Assumption 1, which already imposes population identification. Theorem 3 concerns inference under additional conditions. We have revised the wording to distinguish conditional determination of factors given a fixed and full-rank loading matrix from unique identification of a jointly learned loading--factor decomposition. This precision preserves the central interpretability.
3a. Quantitative economic interpretation and stability
To verify the above interpretability with empirical (quantitative) results, we have extended the characteristic analysis in Appendix A.9 to the newly added Appendix A.17 with confidence intervals for anchored directions, random coordinate transformations, and comparisons across economic anchor sets and graph constructions.
Reply to Reviewer [3/3]
Characteristic structure and persistence. Table A5 gives a median Top-10 characteristic-alignment share of for HeterIPCA, compared with for IPCA and for PCA. The HeterIPCA mean is , with a 95% block-bootstrap interval of . The leading characteristics represent liquidity, trading activity, size, volatility, and market exposure. Eight of the eleven adjacent-month transitions in 2021 retain exactly the same Top-10 set. The other three have Jaccard similarity . Across 600 random rotations of the fitted loading matrix, the alignment profile changes by at most and Top-10 membership remains unchanged. These measurements establish a concentrated and persistent economic structure in the fitted loading subspace.
Economic directions under a fixed convention. We estimate the canonical-correlation transformation and labels over 1957--2008 and apply them unchanged over 2009--2021. Table A13 reports evaluation-period correlations of for the market anchor with 95% interval , and for illiquidity with interval . The third direction combines value and profitability associations, with correlations of and . Across seven estimation windows, the leading market and illiquidity labels recur in six windows, and their top-two ordering is preserved in all seven. The economic labels and signs of the canonical directions are also retained across 500 orthogonal and 500 general invertible coordinate transformations.
External traded factors and graph sensitivity. Table A14 repeats the analysis using Mkt-RF, SMB, HML, RMW, CMA, and momentum, which are external to the factor-network inputs. The alignments and one-to-one matches are determined before 2009 and then held fixed. Across the five graph constructions, the market direction retains its Mkt-RF label, with traded-anchor evaluation correlations of -- and cross-convention absolute correlations of --. The third original direction consistently matches an HML-labeled traded-factor direction, with cross-convention correlations of --. Table A15 shows that for FF49, all six matched score directions have absolute evaluation-period correlations of -- across the two conventions.
These experiments provide economic interpretation at two explicit levels: the rotation-invariant characteristic structure of the loading space, and economically anchored directions in that space. The leading market, liquidity, and value-related associations are supported by direct quantitative evidence, and Tables A13--A15 report the behavior of every direction under the stated conventions.
4. Bottleneck Factor Network and optimization
The Bottleneck Factor Network receives a characteristic-managed portfolio-return representation and passes it through linear layers, layer normalization, and ReLU activations to generate a -dimensional latent-factor representation. The Micro-Graph Network generates , and the Conditional Alpha Network combines representations from micro- and public-instrument branches. The predicted return is built from the systematic component and the conditional alpha component, with squared loss as in Eqs. (7)--(9). We add such details in Section 3.1. Appendix A.4 reports Adam optimization with initial learning rate , weight decay , dropout , gradient clipping at , up to epochs, and early stopping after validation epochs without improvement.
Reply to Reviewer [1/3]
We sincerely thank the reviewer for the constructive assessment and for recognizing the graph-based extension of IPCA, the attempt to provide statistical guarantees, and the favorable economic results. We address theoretical novelty, baseline fairness, interpretability, and implementation in turn.
1. Theoretical contribution relative to Farrell et al. (2025)
The debiasing principle used in our inference analysis employs certain ideas and concepts of the established double/debiased machine learning (DML) framework, including Farrell et al. (2025). However, Theorem 1 (the pseudo-dimension of structured graph convolutional network, SGCN) is a new theoretical result deduced in this work, as only the pseudo-dimension of the ordinary fully connected network (FCN) has been deduced before. From then on, all the theoretical results have to be reformed according to this newly-developed Theorem 1, including Theorems 2 and 3. Therefore, we actually extend the DML framework from FCN to the much more complicated network structure SGCN. Moreover, HeterIPCA (our method) studies a structured and nonlinear instrument-to-loading representation, with separate micro-instrument, public-instrument, and conditional-alpha components, instead of a stand-alone fully connected nuisance estimator.
Our theory also addresses a different statistical function class. In particular, Theorem 1 states the architecture-dependent pseudo-dimension bound
Pdim(SGCN)≤CdQLlog(Qn),
where Q is the number of trainable parameters, L the depth, and n the graph size. Its derivation counts graph-dependent activation operations instead of simply treating the model as an ordinary fully connected network. This is an architecture-specific complexity contribution relevant to graph-based parameter sharing. Theorem 2 then pursues a finite-sample estimation bound for the structured function class. Likewise, Theorem 3 is a DML-type inference extension under its stated regularity and nuisance-rate conditions. These clarifications sharpen the difference between our graph-structured factor framework and the prior fully connected setting.
The revised Appendix A.16 also evaluates the orthogonal correction directly. Across 500 paired replications per sample size, its nominal 95% intervals attain coverage of 97, 98, and 95 (Table A12). This complements the estimation analysis with finite-sample evidence for inference on the average nonlinear loading..
2. Input fairness, public instruments, and the matched MLP
2a. Fairness of the input information and the contribution of public instruments
The revised Appendix A.13 now separates the input configurations from the effect of the architecture. Table A7 specifies each model's inputs: HeterIPCA uses 94 firm characteristics, the market portfolio, and ten public predictors; IPCA uses its 94-characteristic specification; and the portfolio-based and graph-based comparators retain their respective input constructions. We then add the same ten public predictors to IPCA, CA0--CA3, and SDNN and report their predictive and economic performance at K=6 in Table A8.
The augmented baselines improve on the predictive metrics, while HeterIPCA retains the strongest result on all three reported metrics. The most informative comparisons are:
The strongest augmented portfolio-level predictor CA1 increases from 0.735 to 0.762, leaving a 0.798 percentage-point advantage for HeterIPCA. The strongest augmented portfolio strategy, CA0, has net Sharpe 1.63, compared with 2.57 for HeterIPCA. Augmented IPCA achieves stock-level Rr2=0.652, compared with 0.664. Thus, the largest gains are in characteristic-managed portfolio prediction and the economic quality of the return ranking. The matched MLP experiment below provides the additional control in which the complete input set and model capacity are identical.