WorksMathematics
A Hierarchical Group Factor Model with Characteristic-Driven Global and Local Factors
Abstract
We propose a Hierarchical Group Factor Model (HGFM) for jointly estimating global and group-specific factors with characteristic-driven portfolio weights. Both factor types are constructed from contemporaneous returns, producing endogeneity that we address by quasi-maximum likelihood estimation. Conditional on the interaction matrices, we separate the global and local coefficient components and normalize their factor representations. Under a fixed characteristic design and suitable regularity conditions, we establish local consistency of the coefficient estimators and asymptotic normality for fixed blocks of the normalized factor parameters as both panel dimensions diverge. We also propose an information criterion and establish its consistency for jointly selecting the global rank and the common local rank. In an industry-grouped application to S&P 100 constituents, the HGFM achieves higher explanatory power for realized excess returns and a higher portfolio Sharpe ratio than benchmark asset-pricing models.
Introduction
How can we separate common and group-specific dependence when both global and local factors are constructed from observed responses using characteristic-driven weights?
For stock returns grouped by industry, the distinction is between comovement across industries and additional comovement within each industry. Standard factor models summarize common variation in high-dimensional panels (e.g., [7]), whereas group factor models (GFMs) distinguish global factors shared across groups from local factors specific to each group. Grouped heterogeneity also motivates sparse asset-pricing models [14] and group-specific predictability of mutual fund alphas [15]. Characteristic-based portfolios formed across the full stock panel and within individual industries may capture overlapping variation in returns. This motivates jointly specifying and estimating the global and local components within a common return model. We develop a framework in which predetermined asset characteristics generate time-varying weights for both global and local factors.
Andreou, Gagliardini, Ghysels, and Rubin [3] develop inference for group factor models using within-group principal components and cross-group canonical correlations. We use their common and group-specific latent factor structure as a benchmark. Let denote the response variable for the th cross-sectional unit at time in group , where , , and . Define as the vector of response variables for group at time . The group factor model is then specified as follows:
where and are the latent global and local factors, respectively; and are their loading matrices; and is the idiosyncratic error. Thus, affects all groups, whereas affects only group .
The latent-factor specification in (1) leaves the relation between asset characteristics and factor-portfolio weights unspecified. Empirical asset pricing often constructs observable factors by sorting securities on characteristics and forming long–short portfolios, as in the Fama–French factors [17, 18]. The number of such factors has grown large [28], and Feng et al. [23] test whether a new factor adds explanatory power for expected returns beyond the existing set.
We express both types of factors in (1) as linear combinations of contemporaneous returns, with time-varying weights determined by characteristics, and estimate their fixed coefficients jointly. Kelly, Pruitt, and Su [30] use characteristics to parameterize time-varying factor loadings; our characteristics instead determine portfolio weights:
where and . The matrices and are the global and local portfolio weight matrices, respectively, while and are fixed characteristic-coefficient matrices. For group , contains transformed characteristics available before the period- returns are realized, and collects these characteristics across all groups. The portfolio weights vary over time because and vary over time, even though and are fixed. This specification restricts each factor’s portfolio weights to the span of the observed characteristics.
By substituting (2) into the group factor model (1), we propose a Hierarchical Group Factor Model (HGFM):
The global factor in (3) is a portfolio of all assets with weights , while the local factor for group uses only assets in that group, with weights . This linear construction can also represent sorting-based factors (e.g., the Fama–French size and value factors) when the rows of the characteristic matrices encode the stock weights of the constituent portfolios and the coefficient matrices specify their fixed long–short combinations. In our specification, the coefficient matrices are estimated jointly with the other HGFM parameters, allowing the data to determine how predetermined characteristics enter the global and local portfolio weights.
We first separate the global and local coefficient matrices conditional on the interaction matrices, normalize their factor representations, and impose a local information condition for statistical identification. Replacing latent factors with functions of observed returns introduces the outcome on the right-hand side of model (3), which gives rise to endogeneity and generally renders ordinary least squares (OLS) estimators inconsistent. We therefore propose a quasi-maximum likelihood estimator (QMLE) computed by gradient ascent with an Armijo line search.
The asymptotic analysis allows both the cross-sectional dimension, , and the time dimension, , to diverge. Although the coefficient matrices are low-rank, the number of nuisance parameters increases with , complicating the analysis of their asymptotic properties [22]. We establish local consistency for the coefficient estimators and consistency and fixed-block asymptotic normality for the normalized factor parameters. The true ranks of the global and local factors are unknown, and standard eigenvalue-ratio methods (e.g., [31, 1]) cannot be directly applied to select them in the present endogenous model. Following Ando and Bai [2], we propose an information criterion (IC) to jointly select the global rank and the common local rank, which consistently recovers the true ranks under suitable regularity assumptions.
Our model is related to Pu, Fang, Lan, Yu, and Zhang (2025), who estimate reduced-rank spatio-temporal coefficient matrices by QMLE with an Armijo line search. Our approach differs in three respects. First, their specification uses an estimated weight matrix multiplying current and lagged returns, whereas ours constructs global and local factors from contemporaneous returns using characteristic-driven, time-varying weights. Second, we jointly estimate the loading and characteristic-coefficient matrices for the global and local components, allowing the characteristic coefficients of local factors to differ across groups while sharing the global factors. Third, our hierarchical specification requires distinct identification conditions and a corresponding theoretical framework. Conditional on the interaction matrices, cross-group characteristic variation separates the global and local coefficient matrices, and empirical characteristic second moments provide the basis for normalizing their factor representations. We also develop an information criterion to jointly select the global rank and the common local rank. Our asymptotic analysis treats the full characteristic design as fixed in advance and takes probability over the errors.
Related literature. The paper relates to three strands of research: group factor models, proxy-assisted factor estimation, and characteristic-based portfolio construction.
The first strand concerns group factor models. Ando and Bai [2] estimate group membership and group-specific latent factors in panels with observed predictors; our model takes group membership as given and estimates characteristic-driven factor weights. Hu, Li, and Wang (2025) identify latent global factors by maximizing their average correlations with group factors, while Choi, Lin, and Shin [12] use cross-group canonical correlations to select latent global factors, with an application to U.S. industries.
Zhang et al. [40] estimate global and local factors in matrix time series, while Choi and Kim [13] use global and national factors for large volatility matrix estimation. Related factor models also cover three-dimensional panels ([29]), matrix-valued time series ([38]), and tensor observations (e.g., [11, 10]). Factor structures of this kind also underlie high-dimensional covariance and portfolio estimation ([16]). The HGFM jointly estimates global and local coefficient matrices that map predetermined characteristics into time-varying portfolio weights, linking both factor components explicitly to observed characteristics and returns.
The second strand uses observable proxies to estimate latent factors. Fan et al. [19] incorporate observed covariates into smoothed principal component analysis (PCA) to recover latent factors, while Wan, Li, Lu, and Song [37] combine responses and a large set of factor proxies through penalized reduced-rank regression to estimate latent factors. Fan and Liao [20] estimate latent factors using diversified cross-sectional projections, with individual characteristics among the proposed sources of portfolio weight matrices. Related studies examine common factor spaces ([4]), the stability of observable–latent factor relations ([34]), and observable-factor risk premia in the presence of omitted factors ([25]). The HGFM specifies both global and local factors as characteristic-weighted contemporaneous responses and jointly estimates the characteristic coefficients and loadings by QMLE to account for the resulting endogeneity.
The third strand concerns characteristic-based factor construction. Kelly et al. [30] use characteristics to parameterize time-varying factor loadings, and Feng, He, Polson, and Xu [24] jointly learn nonlinear characteristic-based portfolio weights and factor loadings by minimizing aggregate realized pricing errors. Brandt et al. [9] parameterize portfolio weights by firm characteristics and estimate their coefficients by maximizing average investor utility. The HGFM specifies separate linear characteristic-based weighting rules for global and group-specific local factors. For this hierarchical specification, we provide identification conditions and QMLE inference for the weight coefficients and loadings.
Simulation and empirical highlights. We examine the finite-sample accuracy of parameter estimation and rank selection through Monte Carlo simulations. We consider independent idiosyncratic errors under Gaussian, standardized-, and mixture-normal distributions, as well as cross-sectional correlation and heteroskedasticity. Estimation errors generally decline with , including under non-Gaussian errors, and rank-selection accuracy generally improves with . Additional simulations examine robustness to sparse characteristic-driven weights, unequal local ranks, larger cross-sectional dimensions, and alternative penalty constants.
Empirically, we analyze daily returns on S&P 100 constituents from five broad industry groups over 2016–2025. The analysis yields three main findings. First, the global factor assigns greater weight to larger stocks and stocks with higher book-to-market ratios, while the characteristic coefficients of local factors differ across industries. Second, combining global and local factors improves realized-return fit across all industry groups relative to either subset alone and the benchmark models. The combined model also explains average returns better than the benchmarks for the full stock panel, although its advantage varies across industries. Third, the combined portfolio achieves the highest annualized Sharpe ratio and the lowest volatility and maximum drawdown among the portfolios considered. Its Sharpe ratio remains above those of the equal- and value-weighted sample portfolios after transaction costs. Additional results examine sensitivity to industry grouping.
The article is structured as follows. § 2 introduces the HGFM, its identification conditions, and the QMLE. § 4 develops the asymptotic theory and rank-selection procedure. § 5 evaluates finite-sample performance by Monte Carlo simulation. § 6 presents the empirical application, and § 7 concludes. The supplementary material contains proofs and additional simulation and empirical results.
The Model and Methodology
In this section, we formally introduce the HGFM. We begin by specifying the model structure, in which the global and local factors are constructed from observed returns using characteristic-driven weights. We then discuss the identification of the global and local components and their factor representations, and detail the QMLE procedure.
Hierarchical Group Factor Model (HGFM)
By introducing global and local factors constructed from observed returns with characteristic-driven weights, we consider the following hierarchical group factor model:
In matrix form, Eq. (4) can be rewritten as
where denotes the global factor loading matrix, and denotes the local factor loading matrix. Here, forms a block-diagonal matrix from its arguments. Moreover, is the block-diagonal matrix of local characteristic coefficients, and \boldsymbol{\varepsilon}_{t}=\left(\boldsymbol{\varepsilon}_{1,t}^{\mathsf{T},\ldots,\boldsymbol{\varepsilon}_{S,t}^{\mathsf{T}}\right)^{\mathsf{T}}} is the idiosyncratic error vector. Define , , and . Define the time-varying coefficient matrix . Then model (5) is . We refer to as the interaction matrix because it describes contemporaneous dependence among the observed units. When , , , and are fixed, the number of unknown parameters associated with and grows linearly with . This linear growth follows from the fixed dimension of the characteristics, while the low-rank restrictions further constrain the global and local coefficient matrices. Under suitable conditions, the model can be consistently estimated as diverges.
Remark 1 (Relation to spatial models). Our model (5) is related to spatial autoregressive (SAR) models, as both capture contemporaneous dependence. Traditional SAR models rely on a prespecified spatial weight matrix, whereas the HGFM uses predetermined characteristics and estimated low-rank coefficient matrices to construct time-varying interaction matrices. The model decomposes these interactions into global and local components. Setting leaves only within-group interactions, while setting yields a purely global structure. The model therefore represents dependence both across and within groups through observed characteristics.
Model Identifiability
Identification issues commonly arise in latent factor models [5, 6, 8, 39]. The proposed model raises two identification issues: the separation of the global and group-specific components and the factorization of each low-rank coefficient matrix. We address these two issues sequentially. First, we exploit cross-group characteristic variation to separate the global and local coefficient matrices. We then impose a Singular Value Decomposition (SVD) normalization on the low-rank factors of each separated coefficient matrix.
The first identification issue concerns whether the time-invariant global and local coefficient matrices can be separately recovered from the time-varying interaction matrix. To illustrate this issue, let and , and let denote the row block of corresponding to group . The model for group can then be written as . Thus, the cross-group input is governed exclusively by , whereas the own-group input is governed by . Given identification of the time-varying interaction matrix, this structure allows us to distinguish and when the characteristics vary sufficiently both within and outside group .
Unless otherwise stated, we treat the entire characteristic array as a nonstochastic design fixed in advance throughout the theoretical analysis. For each group , let denote the concatenation of , and define the corresponding design matrices by
The following condition ensures that the characteristic variation is nondegenerate both within group and across the groups outside .
Condition 1. There exist constants such that, for every ,
Condition 1 rules out degenerate characteristic directions that would prevent the global and local components from being distinguished. The following proposition states that if two collections generate the same interaction matrices at the design dates, their global and group-specific coefficient matrices coincide.
Proposition 1. Suppose that Condition 1 holds. Consider two collections of coefficient matrices and . For every , if
then and for every .
Under Condition 1 and conditional on the interaction matrices, Proposition 1 separates through cross-group characteristic variation and then through within-group variation, for every .
The second identification issue is the usual rotation indeterminacy of low-rank factor models [6, 35]. For any nonsingular matrices and , the transformations , , , and leave and unchanged. Thus, the factor representations remain unidentified even after the global and local coefficient matrices have been separated. We therefore impose an SVD normalization to identify the factors without changing the corresponding coefficient matrices.
Define the empirical second-moment matrices of the observed characteristics as
Since and , both matrices are positive definite under Condition 1. Using these matrices, the following normalization condition identifies each low-rank factorization up to paired sign changes.
Condition 2. The global and local factor matrices satisfy and for every , where and are diagonal matrices whose diagonal entries are positive, distinct, and arranged in decreasing order.
Condition 2 adopts the standard balanced normalization used in factor models ([6, 8, 39, 35]). The balancing equality fixes relative scale, while diagonality, distinctness, and decreasing ordering remove rotation and permutation indeterminacies, leaving only paired sign changes.
Proposition 2. Suppose that and for every . If both representations satisfy Condition 2, then there exist diagonal sign matrices and , with diagonal entries in , such that
Proposition 2 establishes the uniqueness of the normalized factorization up to paired sign changes. To complete the factorization argument for coefficient matrices with the specified exact ranks and singular-value gaps, we construct a normalized representation by an ordered singular value decomposition.
Proposition 3. Suppose that Condition 1 holds. Assume that and for every . Assume that the positive singular values of are distinct and that, for every , those of are also distinct. Let be its ordered thin singular value decomposition, and define and . For each , let be its ordered thin singular value decomposition, and define and . Then and for every . Moreover, the constructed factors satisfy Condition 2 with and .
For coefficient matrices of the stated exact ranks and with distinct positive singular values, Proposition 3 constructs the normalized factorization and Proposition 2 gives uniqueness up to paired signs. Together with Proposition 1, these results establish the uniqueness of the coefficient decomposition and its normalized factorization for the given interaction matrices.
Remark 2 (Equivariance to the ordering of the observed units). If assets are reordered within the known groups, the responses and columns of the characteristic matrices must be reordered jointly. If the groups are reordered, the group partition must be permuted as well. The rows of the loading matrices follow the same reordering as the assets. The characteristic coefficients and factor series are preserved when the same sign convention is used. This is because jointly reordering returns and characteristics leaves their product unchanged. The normalization therefore does not depend on which assets appear first and does not require anchor assets.
Quasi-maximum Likelihood Estimator
Because appears on both sides of model (5), the contemporaneous regressors depend on the errors, so ordinary least squares is generally inconsistent. The ranks and are treated as known throughout this subsection. We therefore adopt the quasi-maximum likelihood approach of Lee [32]. Let collect the factor parameters before normalization, and let , , and . The associated coefficient parameter is . Let be the interaction matrix in model (5) generated by and , and define . Up to an additive constant independent of the parameters, the sample-size-normalized negative quasi-log-likelihood is
Let denote the set of whose coefficient products and variance vector satisfy the compactness, rank, and stability restrictions in Condition 3. Estimation proceeds by minimizing over this parameter space.
The variance parameters can be concentrated out of the criterion. For fixed , the variance first-order conditions give
where denotes the th canonical vector in . Substituting (7) into (6), multiplying by , and omitting the resulting constant gives the concentrated quasi-log-likelihood
Following He et al. (2023) and Pu et al. (2025), we maximize over the factor matrices using a gradient-based algorithm with an Armijo line search. Denote the retained, not-yet-normalized factor matrices by . Neither singular-value truncation nor the SVD normalization is imposed during the iterations. In implementation, the optimization is repeated from multiple full-rank initial values, and the candidate attaining the largest value of is retained. The corresponding log-variance estimates are .
After optimization, define and . We then apply the post-estimation normalization in Proposition 3 to these estimated coefficient matrices, using and for the global and local components, respectively, obtaining , , , and . By construction, and for every . Consequently, the normalization leaves the estimated coefficient matrices, variance estimates, and quasi-likelihood value unchanged while selecting an identified factor representation.
Algorithm 1: Gradient algorithm with Armijo line search
Input: Group returns ; predetermined characteristics ; ranks ; initial values , where ; maximum numbers of iterations ; scale factor ; tolerance ; Armijo parameter .
Output: .
for do
for do
Compute from (7), , and ;
if then
Mark the current estimate as converged and break;
end
for do
Set and ;
if then
Set and break;
end
end
end
end
Among the admissible converged terminal estimates, retain the one with the largest profile likelihood; denote its factors by ;
Apply the normalization in Proposition 3 to the retained coefficient matrices and , using and , respectively, and denote the normalized factors by ;
return ;
Remark 3 (Scope of the numerical guarantee). Algorithm 1 produces numerical candidate estimates from initial values. The Armijo line search increases the objective along accepted updates, but neither the line search nor multiple starts guarantees a global optimum. The theory below establishes the existence and rate of a local solution in a neighborhood of the identified parameter.
Theoretical Results
Theoretical Results
Under the stated conditions, we establish local consistency for the coefficient and log-variance estimators, transfer this rate to the normalized factor representation, and derive asymptotic normality for prespecified loading rows and characteristic-coefficient columns. We then establish consistency of joint global and common local rank selection under suitable penalty rates.
Average Consistency
Before establishing the average consistency of the proposed QMLE, we introduce some notation. For a generic matrix , denotes its column-wise vectorization. Let and denote the minimum and maximum eigenvalues of a symmetric matrix , respectively. In addition, denotes the induced matrix -norm for , and denotes the Frobenius norm.
Let denote the coefficient and log-variance parameters, and let denote the column-wise vectorization of . The corresponding scaling matrix is . For a vector with the same dimension as , set . For two coefficient-space parameters and , use the same notation for their vectorized difference:
Equivalently, .
For average consistency, we use the coefficient parameter . Define , where is the interaction matrix in (5) written in terms of and .
Since depends on the direct factors only through and , its equivalent coefficient-space form is
The admissible coefficient space is the compact, rank-restricted, and stable set specified in Condition 3 below. Define the population criterion by on the admissible coefficient space. Let denote the data-generating parameter; it minimizes the population criterion.
We next state the regularity conditions.
Condition 3. Let be a compact parameter space for , with and for each . There exist constants , such that, for all ,
Moreover, is nonsingular for every , and
Condition 4. The true coefficient matrices have ranks and , and is uniformly interior to the non-rank restrictions in Condition 3. At the normalized factor representation, let and write . For each , let and write . There exist constants and , independent of , such that every diagonal entry of these matrices lies in and every adjacent difference within each matrix is at least .
Condition 5. The number of groups is fixed, and there exist constants such that for every .
Condition 6. For some constant , the characteristic array satisfies
The error vectors are independent over time, and their entries are mutually independent within each date. For some and every , , , and , where and .
Condition 7. Let denote the population Hessian with respect to the vectorized coefficient and log-variance parameters. There exist constants , independent of and , such that,
Condition 3 ensures that the simultaneous system remains uniformly nonsingular and that the associated spatial multipliers are uniformly bounded. Such stability restrictions are standard in the asymptotic analysis of spatial QMLE ([32, 33]).
Condition 4 ensures a regular interior parameter and stable SVD normalization. Condition 5 prevents any local block from becoming asymptotically negligible relative to the full cross section. Condition 6 assumes independence over time and across units for the errors. Its moment and independence assumptions control the linear and quadratic forms within each period and support a Lindeberg–Feller central limit theorem. Condition 7 requires the scaled population Hessian to be uniformly positive definite and bounded. It therefore provides local identification and ensures the uniform invertibility needed for the quadratic expansion of the quasi-likelihood. Related curvature conditions are commonly imposed in the inferential theory for high-dimensional generalized factor models ([39]).
We now establish average consistency for the separately identified global and local coefficient components.
Theorem 1. Suppose that Conditions 1, 3, 5, 6, and 7 hold, and let with . Then, with probability approaching one, admits a local minimizer $\widehat{\eta}\in
Asymptotic Normality
satisfying
Theorem 1 implies and for every , while . Thus, the global and each local coefficient matrix converge at the root- rate in Frobenius norm, and the average squared error of the log-variance estimates is of order .
We next transfer the coefficient-space convergence rates to the normalized factor representation. For this purpose, write , and let denote the vector obtained by stacking these blocks in the displayed order, with each matrix vectorized columnwise. We use and to denote the normalized true and estimated factor representations, respectively, while a generic is not required to satisfy the normalization conditions.
To account for the different dimensions of the loading and characteristic-coefficient blocks, define the scaled factor norm by
Corollary 1. Suppose that Conditions 1–7 hold and . Let be obtained from by applying the SVD normalization in Proposition 3 to the estimated coefficient matrices while leaving unchanged. After paired sign alignment,
Corollary 1 shows that the coefficient-space convergence rate is preserved by the post-estimation SVD normalization. Specifically, the normalized global and local loading matrices, characteristic-coefficient matrices, and log-variance parameters are jointly consistent at the root- rate under the scaled factor norm.
Asymptotic Normality
For individual asymptotic normality, consider a prespecified loading row or characteristic-coefficient column as . Specifically, for , for , , and . Define
Accordingly, for or and for or . We use and for the corresponding blocks of the sign-aligned local estimator in Corollary 1 and of the normalized truth.
For each fixed block , let denote the covariance matrix of the leading term in the first-order expansion of . Its explicit form is given in Appendix S3.
Condition 8. For every fixed block in (13), there exists a positive-definite matrix with . As , for or , for , and for .
Following the fixed-index formulation of Ouyang et al. [35], Condition 8 separately assumes that the appropriately scaled covariance of each fixed block has a uniformly positive definite limit. The scaling may differ between loading rows and characteristic-coefficient columns.
Theorem 2. Suppose that Conditions – hold. If with , then for every fixed block in (13),
Theorem 2 gives the asymptotic normal distribution for each prespecified loading row or characteristic-coefficient column, standardized by its own covariance.
Remark 4. The average-consistency result only requires , whereas Theorem 2 is stated under the stronger condition . The reason is that consistency only requires the estimation error to vanish under the scaled parameter norm, while fixed-block asymptotic normality further requires the remainder in the first-order linear representation to be negligible after covariance standardization. The covariance matrix accounts for the different fluctuation scales across the loading and characteristic-coefficient blocks. Under this normalization, the Hessian-replacement remainder used in the proof is uniformly bounded by over the four types of fixed blocks. Hence is sufficient for this remainder to be and yields a unified growth condition for Theorem 2.
Rank Selection
The preceding results assume that the global and local ranks are known. We now consider their joint selection using an information criterion. Let denote the true rank pair, and let be the candidate set, where is fixed. For each , let consist of , where , , , , , and . Their coefficient products and variance parameters are required to satisfy the compactness and stability restrictions in Condition 3, with replaced by . Write .
For later comparison across candidate ranks, let and , and write if and . For an underfitted candidate , let denote a population minimizer of over . For an exact-fit or overfitted candidate , set , which belongs to and minimizes the population criterion.
For each , let denote the rank-specific local QMLE, and set . We define the information criterion by
Algorithm 2 implements rank selection using the numerical estimates from Algorithm 1.
For exact-fit and overfitted candidates, use the local minimizer near constructed in Appendix S4.
Condition 9. There is a constant , independent of , such that every underfitted candidate satisfies
Condition 9 requires the minimum population criterion under every underfitted rank specification to exceed by at least . The minimization is joint over the coefficient matrices and the log-variance vector in , with denoting a population minimizer. For exact-fit and overfitted candidates, local quadratic growth around follows from Condition 7.
Remark 5 (Verification of Condition 9). Proposition 5 in Section S4.2 of the supplementary material gives a sufficient condition for Condition 9. It requires the covariance matrices at each underfitted population minimizer to remain separated from the true covariance matrices, as measured by their time-averaged squared Frobenius distance. Under the stability bounds, this gives the required population loss of order . Section S5.6 provides numerical evidence for this loss in the designs considered.
Theorem 3. Let be defined by (14). Suppose that Condition 1, Conditions 3–7, and Condition 9 hold. If , , , and , then .
The restrictions and balance the effects of overfitting and underfitting. Specifically, ensures that the rank penalty dominates the reduction in the sample criterion due to overfitting, whereas makes the penalty asymptotically negligible relative to the order- population loss from underfitting. We set , which satisfies these restrictions when . In the simulation study, we use as the baseline constant, which gives reasonable rank-selection accuracy in the designs considered. To assess robustness to this choice, we also conduct a sensitivity analysis over a range of constants in Appendix S5.5.
Algorithm 2: Rank selection: minimization of
Input: Maximum rank: .
Output: .
for and do
Run Algorithm 1 with ranks and calculate ;
end
Select and that minimize in (14);
return ;
Simulation Studies
We use Monte Carlo simulations to evaluate the finite-sample performance of QMLE and the accuracy of global and local rank selection. Two distinct simulation environments are considered. The first model adopts independent, idiosyncratic errors with various tail behaviors, whereas the second incorporates cross-sectional dependence and heteroskedasticity commonly observed in economic and financial applications.
Two Simulation Settings
We construct the data-generating processes based on the model specified in (5). To obtain coefficient matrices of the specified ranks and a stable simultaneous system, we generate the loading matrices and errors as follows.
Case 1. We first generate matrices and . Let and be orthonormal bases obtained from independent standard Gaussian matrices. We consider . For , let be the matrix with entry 1; for , let . We construct and . Thus, the global characteristic coefficient matrix has rank . Similarly, for group , we independently generate orthonormal bases and . Define in the same way using . We construct and , so the local characteristic coefficient matrix has rank . At each , we generate a characteristic matrix whose entries are independent draws from , where denotes the number of characteristics, with in the simulations. The global component uses , and the local component for group uses the corresponding columns, denoted by . The resulting time-varying interaction matrix is
To set the scale of the coefficients, we generate a separate characteristic path of length and multiply , , , and by a common constant so that along this path, and generate , .
For the noise term , we first draw and set . To examine sensitivity to the error distribution, we generate under three specifications: (1) the elements of are independent standard normal variables; (2) the elements of are independent variables divided by ; and (3) the elements of are independent draws from . In all cases, given , the errors have mean zero and covariance matrix .
Case 2. The second experiment examines estimation under blockwise cross-sectional dependence and heteroskedasticity. The coefficient and characteristic processes are generated as in Case 1, but the covariance matrix is different. Following the design in Fan et al. [21], we initialize and partition the first units into blocks of size four. We randomly select of these blocks and replace each selected block by the correlation matrix whose entry is for . Any remaining units are uncorrelated with one another and with all other units. To incorporate cross-sectional heteroskedasticity, we draw , define , and set . This design assesses robustness when the errors exhibit cross-sectional correlation and heteroskedasticity.
We set the number of groups to with equal group sizes . The true ranks for global and local factors, , vary across the set , yielding four distinct combinations. We consider a range of sample sizes with and , resulting in nine pairs. These settings are similar to the standard data sets used in economics and include combinations similar to the cross-sectional and time-series dimensions of our industry portfolio application. For each setting, we replicate the experiment 200 times, keeping the data-generating coefficient matrices and error variances fixed across replications.
For the QMLE procedure in Algorithm 1, we use the normalized objective and set , , , , and . For each random start, we generate initial values by drawing auxiliary loading and characteristic-coefficient entries independently from and , respectively. An SVD of their products, followed by a common rescaling, yields , and, for each group, , . Among the converged estimates, we retain the one with the highest concentrated quasi-log-likelihood.
Simulation Results for QMLE Estimation
To quantify the estimation accuracy, we report the means (Mean) and standard deviations (SD) of the estimation errors across 200 replications. We use the Frobenius norm for matrices and the Euclidean norm for vectors, scaling each by its respective dimensions to facilitate comparison. Specifically, the estimation errors for the model parameters are computed as follows: , , , , , , and . Here, , , , and denote the corresponding normalized true parameters. We also report the response-fit error , reported in the columns labeled Fit error. In these metrics, lower values indicate more accurate estimation.
Tables 1 and 2 summarize the simulation results for Cases 1 and 2 under normal errors, respectively. Three patterns emerge. First, for fixed and ranks, the mean estimation errors and response-fit errors decrease as increases, while their standard deviations generally decline. Second, comparing the two cases, coefficient matrix errors are similar, whereas variance estimation and response-fit errors are larger under the dependent errors in Case 2. Nevertheless, both errors decrease with . Third, for fixed and ranks, increasing generally reduces coefficient matrix errors, although its effects on individual factor parameter errors are less uniform.
The supplementary material examines robustness to non-Gaussian errors (Section S5.1), unequal local ranks (Section S5.2), sparse portfolio weights (Section S5.3), and larger cross-sectional dimensions (Section S5.4). Overall, the estimation errors exhibit broadly similar patterns across these settings, supporting the robustness of the estimator’s finite-sample performance.
| B | U | Fit error | |||||||
| 60 | 1000 | 0.0047(0.0004) | 0.0442(0.0043) | 0.0100(0.0034) | 0.0081(0.0008) | 0.0562(0.0058) | 0.0202(0.0053) | 0.0571(0.0052) | 0.0521(0.0035) |
| 1500 | 0.0038(0.0003) | 0.0363(0.0035) | 0.0073(0.0032) | 0.0066(0.0006) | 0.0465(0.0046) | 0.0160(0.0037) | 0.0467(0.0043) | 0.0424(0.0027) | |
| 2000 | 0.0032(0.0003) | 0.0314(0.0029) | 0.0060(0.0025) | 0.0057(0.0006) | 0.0405(0.0042) | 0.0133(0.0033) | 0.0406(0.0036) | 0.0365(0.0025) | |
| 90 | 1000 | 0.0039(0.0003) | 0.0444(0.0036) | 0.0097(0.0029) | 0.0068(0.0006) | 0.0579(0.0051) | 0.0193(0.0045) | 0.0560(0.0043) | 0.0521(0.0030) |
| 1500 | 0.0031(0.0003) | 0.0369(0.0029) | 0.0072(0.0024) | 0.0055(0.0004) | 0.0477(0.0038) | 0.0147(0.0036) | 0.0455(0.0036) | 0.0422(0.0022) | |
| 2000 | 0.0027(0.0002) | 0.0320(0.0025) | 0.0058(0.0019) | 0.0048(0.0004) | 0.0418(0.0034) | 0.0122(0.0031) | 0.0394(0.0031) | 0.0364(0.0019) | |
| 120 | 1000 | 0.0032(0.0002) | 0.0402(0.0026) | 0.0081(0.0024) | 0.0057(0.0004) | 0.0524(0.0034) | 0.0164(0.0036) | 0.0565(0.0033) | 0.0514(0.0022) |
| 1500 | 0.0026(0.0002) | 0.0332(0.0021) | 0.0058(0.0020) | 0.0046(0.0003) | 0.0433(0.0030) | 0.0121(0.0030) | 0.0461(0.0028) | 0.0416(0.0019) | |
| 2000 | 0.0022(0.0001) | 0.0290(0.0018) | 0.0046(0.0016) | 0.0039(0.0003) | 0.0377(0.0025) | 0.0100(0.0024) | 0.0399(0.0028) | 0.0359(0.0016) | |
| \multicolumn{10}{c}{Panel A: } | |||||||||
| 60 | 1000 | 0.0047(0.0005) | 0.0448(0.0045) | 0.0105(0.0036) | 0.0110(0.0009) | 0.0657(0.0058) | 0.0329(0.0092) | 0.0564(0.0053) | 0.0633(0.0038) |
| 1500 | 0.0038(0.0004) | 0.0369(0.0036) | 0.0080(0.0032) | 0.0089(0.0006) | 0.0542(0.0040) | 0.0247(0.0059) | 0.0461(0.0044) | 0.0514(0.0029) | |
| 2000 | 0.0033(0.0003) | 0.0319(0.0030) | 0.0067(0.0027) | 0.0077(0.0005) | 0.0471(0.0034) | 0.0203(0.0047) | 0.0400(0.0037) | 0.0443(0.0025) | |
| 90 | 1000 | 0.0038(0.0003) | 0.0420(0.0036) | 0.0092(0.0030) | 0.0089(0.0005) | 0.0610(0.0039) | 0.0286(0.0077) | 0.0565(0.0044) | 0.0634(0.0032) |
| 1500 | 0.0031(0.0003) | 0.0346(0.0028) | 0.0070(0.0025) | 0.0072(0.0004) | 0.0503(0.0031) | 0.0203(0.0046) | 0.0458(0.0036) | 0.0513(0.0024) | |
| 2000 | 0.0026(0.0002) | 0.0299(0.0023) | 0.0056(0.0020) | 0.0062(0.0003) | 0.0443(0.0026) | 0.0170(0.0040) | 0.0396(0.0030) | 0.0443(0.0021) | |
| 120 | 1000 | 0.0034(0.0002) | 0.0436(0.0028) | 0.0099(0.0027) | 0.0079(0.0004) | 0.0626(0.0034) | 0.0301(0.0071) | 0.0566(0.0033) | 0.0633(0.0023) |
| 1500 | 0.0027(0.0002) | 0.0362(0.0024) | 0.0071(0.0023) | 0.0064(0.0003) | 0.0526(0.0030) | 0.0222(0.0054) | 0.0462(0.0029) | 0.0514(0.0020) | |
| 2000 | 0.0024(0.0002) | 0.0317(0.0020) | 0.0058(0.0020) | 0.0055(0.0003) | 0.0460(0.0023) | 0.0176(0.0040) | 0.0400(0.0029) | 0.0442(0.0016) | |
| \multicolumn{10}{c}{Panel B: , } | |||||||||
| 60 | 1000 | 0.0064(0.0004) | 0.0517(0.0033) | 0.0168(0.0056) | 0.0086(0.0008) | 0.0614(0.0058) | 0.0262(0.0075) | 0.0572(0.0053) | 0.0639(0.0035) |
| 1500 | 0.0051(0.0003) | 0.0432(0.0029) | 0.0123(0.0038) | 0.0069(0.0007) | 0.0504(0.0051) | 0.0207(0.0058) | 0.0466(0.0043) | 0.0518(0.0028) | |
| 2000 | 0.0044(0.0003) | 0.0378(0.0023) | 0.0100(0.0031) | 0.0059(0.0006) | 0.0437(0.0044) | 0.0174(0.0046) | 0.0406(0.0037) | 0.0447(0.0023) | |
| 90 | 1000 | 0.0051(0.0003) | 0.0458(0.0027) | 0.0130(0.0043) | 0.0068(0.0005) | 0.0537(0.0043) | 0.0184(0.0043) | 0.0557(0.0043) | 0.0629(0.0030) |
| 1500 | 0.0041(0.0002) | 0.0382(0.0022) | 0.0092(0.0027) | 0.0055(0.0005) | 0.0443(0.0032) | 0.0143(0.0034) | 0.0453(0.0036) | 0.0509(0.0023) | |
| 2000 | 0.0035(0.0002) | 0.0335(0.0019) | 0.0074(0.0020) | 0.0047(0.0004) | 0.0384(0.0029) | 0.0121(0.0029) | 0.0392(0.0031) | 0.0439(0.0019) | |
| 120 | 1000 | 0.0044(0.0002) | 0.0441(0.0021) | 0.0127(0.0033) | 0.0057(0.0004) | 0.0522(0.0034) | 0.0173(0.0041) | 0.0562(0.0032) | 0.0624(0.0024) |
| 1500 | 0.0035(0.0002) | 0.0369(0.0017) | 0.0091(0.0021) | 0.0046(0.0003) | 0.0433(0.0028) | 0.0133(0.0030) | 0.0459(0.0029) | 0.0507(0.0020) | |
| 2000 | 0.0030(0.0001) | 0.0326(0.0014) | 0.0074(0.0021) | 0.0040(0.0003) | 0.0379(0.0024) | 0.0109(0.0026) | 0.0397(0.0028) | 0.0438(0.0017) | |
| \multicolumn{10}{c}{Panel C: , } | |||||||||
| 60 | 1000 | 0.0068(0.0005) | 0.0546(0.0037) | 0.0183(0.0058) | 0.0121(0.0009) | 0.0728(0.0065) | 0.0395(0.0106) | 0.0554(0.0050) | 0.0727(0.0035) |
| 1500 | 0.0055(0.0004) | 0.0457(0.0033) | 0.0135(0.0043) | 0.0097(0.0007) | 0.0600(0.0050) | 0.0300(0.0081) | 0.0452(0.0042) | 0.0589(0.0029) | |
| 2000 | 0.0048(0.0003) | 0.0399(0.0028) | 0.0107(0.0035) | 0.0084(0.0007) | 0.0524(0.0043) | 0.0247(0.0069) | 0.0393(0.0038) | 0.0508(0.0026) | |
| 90 | 1000 | 0.0055(0.0003) | 0.0503(0.0029) | 0.0162(0.0047) | 0.0097(0.0006) | 0.0677(0.0047) | 0.0347(0.0090) | 0.0568(0.0044) | 0.0737(0.0030) |
| 1500 | 0.0044(0.0002) | 0.0421(0.0022) | 0.0116(0.0032) | 0.0078(0.0004) | 0.0555(0.0035) | 0.0248(0.0066) | 0.0461(0.0036) | 0.0595(0.0022) | |
| 2000 | 0.0038(0.0002) | 0.0367(0.0020) | 0.0090(0.0025) | 0.0067(0.0004) | 0.0484(0.0028) | 0.0199(0.0050) | 0.0393(0.0030) | 0.0512(0.0020) | |
| 120 | 1000 | 0.0048(0.0002) | 0.0487(0.0022) | 0.0157(0.0036) | 0.0085(0.0004) | 0.0655(0.0036) | 0.0325(0.0081) | 0.0556(0.0031) | 0.0725(0.0021) |
| 1500 | 0.0039(0.0002) | 0.0410(0.0018) | 0.0112(0.0025) | 0.0069(0.0003) | 0.0551(0.0030) | 0.0242(0.0062) | 0.0454(0.0028) | 0.0587(0.0020) | |
| 2000 | 0.0033(0.0002) | 0.0363(0.0017) | 0.0091(0.0023) | 0.0059(0.0003) | 0.0482(0.0025) | 0.0192(0.0048) | 0.0393(0.0027) | 0.0506(0.0017) | |
| \multicolumn{10}{c}{Panel D: } |
Table 2. Means and standard deviations (in parentheses) for Case 2 under normal errors. Notes: This table reports mean estimation errors and standard deviations (in parentheses) for Case 2 under normal errors. The table format follows Table 1.
| B | U | Fit error | |||||||
| 60 | 1000 | 0.0046(0.0004) | 0.0438(0.0040) | 0.0099(0.0035) | 0.0081(0.0008) | 0.0560(0.0058) | 0.0201(0.0049) | 0.0350(0.0032) | 0.0404(0.0027) |
| 1500 | 0.0038(0.0004) | 0.0364(0.0034) | 0.0078(0.0029) | 0.0066(0.0006) | 0.0463(0.0045) | 0.0161(0.0040) | 0.0286(0.0027) | 0.0331(0.0020) | |
| 2000 | 0.0033(0.0003) | 0.0316(0.0029) | 0.0063(0.0026) | 0.0057(0.0005) | 0.0404(0.0037) | 0.0135(0.0035) | 0.0247(0.0024) | 0.0285(0.0017) | |
| 90 | 1000 | 0.0038(0.0003) | 0.0442(0.0036) | 0.0095(0.0028) | 0.0067(0.0005) | 0.0573(0.0042) | 0.0191(0.0046) | 0.0341(0.0028) | 0.0399(0.0022) |
| 1500 | 0.0031(0.0002) | 0.0364(0.0030) | 0.0071(0.0021) | 0.0055(0.0004) | 0.0475(0.0034) | 0.0143(0.0033) | 0.0278(0.0022) | 0.0323(0.0017) | |
| 2000 | 0.0027(0.0002) | 0.0317(0.0025) | 0.0058(0.0023) | 0.0047(0.0004) | 0.0412(0.0033) | 0.0117(0.0027) | 0.0242(0.0019) | 0.0278(0.0015) | |
| 120 | 1000 | 0.0032(0.0002) | 0.0395(0.0028) | 0.0082(0.0022) | 0.0056(0.0004) | 0.0516(0.0032) | 0.0160(0.0036) | 0.0345(0.0022) | 0.0395(0.0019) |
| 1500 | 0.0026(0.0002) | 0.0330(0.0022) | 0.0059(0.0019) | 0.0045(0.0003) | 0.0427(0.0028) | 0.0122(0.0030) | 0.0281(0.0019) | 0.0321(0.0013) | |
| 2000 | 0.0022(0.0001) | 0.0290(0.0019) | 0.0047(0.0016) | 0.0039(0.0003) | 0.0375(0.0025) | 0.0103(0.0026) | 0.0242(0.0018) | 0.0278(0.0013) | |
| 60 | 1000 | 0.0047(0.0005) | 0.0447(0.0041) | 0.0106(0.0043) | 0.0110(0.0008) | 0.0652(0.0051) | 0.0323(0.0088) | 0.0343(0.0033) | 0.0490(0.0027) |
| 1500 | 0.0039(0.0004) | 0.0368(0.0035) | 0.0082(0.0035) | 0.0089(0.0006) | 0.0543(0.0039) | 0.0241(0.0068) | 0.0279(0.0027) | 0.0399(0.0020) | |
| 2000 | 0.0033(0.0003) | 0.0320(0.0032) | 0.0069(0.0030) | 0.0077(0.0006) | 0.0475(0.0037) | 0.0203(0.0054) | 0.0241(0.0023) | 0.0345(0.0019) | |
| 90 | 1000 | 0.0037(0.0003) | 0.0417(0.0036) | 0.0091(0.0030) | 0.0088(0.0005) | 0.0607(0.0039) | 0.0284(0.0067) | 0.0344(0.0027) | 0.0488(0.0022) |
| 1500 | 0.0030(0.0003) | 0.0342(0.0031) | 0.0069(0.0023) | 0.0071(0.0004) | 0.0506(0.0036) | 0.0209(0.0057) | 0.0279(0.0021) | 0.0395(0.0017) | |
| 2000 | 0.0026(0.0002) | 0.0298(0.0025) | 0.0057(0.0022) | 0.0061(0.0003) | 0.0440(0.0024) | 0.0168(0.0039) | 0.0244(0.0018) | 0.0340(0.0014) | |
| 120 | 1000 | 0.0034(0.0002) | 0.0431(0.0030) | 0.0101(0.0028) | 0.0078(0.0004) | 0.0623(0.0037) | 0.0298(0.0079) | 0.0345(0.0020) | 0.0488(0.0018) |
| 1500 | 0.0027(0.0002) | 0.0361(0.0024) | 0.0071(0.0023) | 0.0063(0.0003) | 0.0520(0.0029) | 0.0210(0.0051) | 0.0280(0.0019) | 0.0395(0.0014) | |
| 2000 | 0.0024(0.0001) | 0.0316(0.0020) | 0.0056(0.0019) | 0.0055(0.0003) | 0.0461(0.0026) | 0.0172(0.0049) | 0.0241(0.0018) | 0.0342(0.0012) | |
| 60 | 1000 | 0.0063(0.0004) | 0.0515(0.0034) | 0.0162(0.0043) | 0.0085(0.0007) | 0.0612(0.0058) | 0.0255(0.0064) | 0.0350(0.0031) | 0.0495(0.0024) |
| 1500 | 0.0052(0.0004) | 0.0432(0.0032) | 0.0123(0.0037) | 0.0069(0.0006) | 0.0505(0.0043) | 0.0204(0.0053) | 0.0285(0.0028) | 0.0402(0.0020) | |
| 2000 | 0.0044(0.0003) | 0.0378(0.0028) | 0.0100(0.0032) | 0.0059(0.0005) | 0.0436(0.0039) | 0.0169(0.0043) | 0.0246(0.0023) | 0.0347(0.0019) | |
| 90 | 1000 | 0.0051(0.0003) | 0.0456(0.0027) | 0.0131(0.0039) | 0.0067(0.0005) | 0.0537(0.0038) | 0.0185(0.0046) | 0.0339(0.0027) | 0.0483(0.0020) |
| 1500 | 0.0041(0.0002) | 0.0382(0.0021) | 0.0096(0.0029) | 0.0054(0.0004) | 0.0441(0.0034) | 0.0146(0.0036) | 0.0275(0.0021) | 0.0392(0.0017) | |
| 2000 | 0.0035(0.0002) | 0.0334(0.0018) | 0.0076(0.0024) | 0.0047(0.0003) | 0.0380(0.0028) | 0.0123(0.0032) | 0.0240(0.0019) | 0.0337(0.0015) | |
| 120 | 1000 | 0.0043(0.0002) | 0.0435(0.0021) | 0.0124(0.0029) | 0.0057(0.0004) | 0.0521(0.0035) | 0.0173(0.0038) | 0.0341(0.0021) | 0.0479(0.0017) |
| 1500 | 0.0035(0.0002) | 0.0367(0.0017) | 0.0091(0.0020) | 0.0046(0.0003) | 0.0428(0.0030) | 0.0129(0.0029) | 0.0277(0.0018) | 0.0389(0.0014) | |
| 2000 | 0.0030(0.0001) | 0.0324(0.0016) | 0.0071(0.0016) | 0.0039(0.0003) | 0.0374(0.0025) | 0.0105(0.0026) | 0.0239(0.0017) | 0.0336(0.0012) | |
| 60 | 1000 | 0.0068(0.0005) | 0.0545(0.0037) | 0.0182(0.0054) | 0.0120(0.0008) | 0.0720(0.0058) | 0.0391(0.0100) | 0.0332(0.0033) | 0.0557(0.0026) |
| 1500 | 0.0055(0.0004) | 0.0455(0.0032) | 0.0134(0.0048) | 0.0096(0.0006) | 0.0593(0.0047) | 0.0292(0.0077) | 0.0271(0.0028) | 0.0451(0.0021) | |
| 2000 | 0.0047(0.0003) | 0.0398(0.0029) | 0.0105(0.0038) | 0.0083(0.0006) | 0.0519(0.0041) | 0.0239(0.0063) | 0.0233(0.0023) | 0.0388(0.0018) | |
| 90 | 1000 | 0.0055(0.0003) | 0.0499(0.0029) | 0.0160(0.0044) | 0.0090(0.0006) | 0.0664(0.0045) | 0.0327(0.0080) | 0.0349(0.0027) | 0.0568(0.0022) |
| 1500 | 0.0044(0.0002) | 0.0419(0.0024) | 0.0119(0.0033) | 0.0077(0.0004) | 0.0552(0.0036) | 0.0236(0.0061) | 0.0284(0.0022) | 0.0461(0.0017) | |
| 2000 | 0.0038(0.0002) | 0.0365(0.0021) | 0.0091(0.0026) | 0.0066(0.0004) | 0.0479(0.0028) | 0.0193(0.0043) | 0.0247(0.0020) | 0.0395(0.0014) | |
| 120 | 1000 | 0.0047(0.0002) | 0.0482(0.0023) | 0.0158(0.0036) | 0.0084(0.0004) | 0.0652(0.0039) | 0.0323(0.0084) | 0.0334(0.0020) | 0.0554(0.0018) |
| 1500 | 0.0038(0.0002) | 0.0406(0.0021) | 0.0116(0.0028) | 0.0068(0.0004) | 0.0544(0.0032) | 0.0231(0.0062) | 0.0272(0.0018) | 0.0449(0.0014) | |
| 2000 | 0.0033(0.0002) | 0.0360(0.0018) | 0.0091(0.0022) | 0.0058(0.0003) | 0.0476(0.0027) | 0.0184(0.0045) | 0.0234(0.0017) | 0.0386(0.0013) |
Table 1. Means and standard deviations (in parentheses) for Case 1 under normal errors. Notes: This table reports mean estimation errors and standard deviations (in parentheses) for Case 1 under normal errors. Columns report errors for the global parameters , B, , local parameters , U, , error variances , and response-fit error (Fit error). The number of global () and local () factors was chosen from , yielding four combinations detailed in Panels A to D.
Simulation Results for Rank Selection
We then evaluate the performance of the IC proposed in (14) for determining the numbers of global and local factors. Table 3 reports the probabilities of correctly selecting each rank and both ranks jointly using the IC in Eq. (14) under normal errors. First, for most rank combinations, the joint selection probabilities approach one as increases for a fixed . Selection is more difficult when both true ranks are larger, and the improvement with is not always monotone. Second, larger cross-sectional dimensions can reduce selection accuracy in smaller samples. Comparing Case 2 with Case 1, however, the presence of cross-sectional dependence and heteroskedasticity does not systematically reduce selection accuracy.
| Panel A: Case 1 | |||||||||||||
| 60 | 1000 | 1.00 | 0.98 | 0.98 | 0.99 | 0.99 | 0.98 | 0.95 | 0.93 | 0.90 | 0.96 | 0.91 | 0.87 |
| 1500 | 1.00 | 1.00 | 1.00 | 1.00 | 0.99 | 0.99 | 0.96 | 1.00 | 0.96 | 0.91 | 0.89 | 0.80 | |
| 2000 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 0.98 | 1.00 | 0.98 | 0.98 | 0.94 | 0.92 | |
| 90 | 1000 | 0.91 | 0.86 | 0.79 | 0.88 | 1.00 | 0.88 | 1.00 | 0.89 | 0.89 | 0.97 | 0.97 | 0.94 |
| 1500 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 0.99 | 0.99 | |
| 2000 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | |
| 120 | 1000 | 0.81 | 0.68 | 0.54 | 0.79 | 1.00 | 0.79 | 0.94 | 0.61 | 0.57 | 0.92 | 0.99 | 0.91 |
| 1500 | 1.00 | 0.99 | 0.99 | 1.00 | 0.99 | 0.99 | 1.00 | 1.00 | 1.00 | 0.98 | 1.00 | 0.98 | |
| 2000 | 1.00 | 1.00 | 1.00 | 1.00 | 0.98 | 0.98 | 1.00 | 1.00 | 1.00 | 0.95 | 1.00 | 0.95 | |
| Panel B: Case 2 | |||||||||||||
| 60 | 1000 | 0.99 | 0.98 | 0.97 | 0.99 | 0.98 | 0.97 | 0.89 | 0.99 | 0.88 | 0.89 | 0.94 | 0.83 |
| 1500 | 1.00 | 1.00 | 1.00 | 1.00 | 0.99 | 0.99 | 0.93 | 1.00 | 0.93 | 0.91 | 0.91 | 0.82 | |
| 2000 | 1.00 | 1.00 | 1.00 | 1.00 | 0.99 | 0.99 | 0.95 | 1.00 | 0.95 | 0.92 | 0.93 | 0.85 | |
| 90 | 1000 | 0.96 | 0.88 | 0.85 | 0.94 | 1.00 | 0.94 | 0.97 | 0.87 | 0.85 | 0.99 | 0.99 | 0.98 |
| 1500 | 1.00 | 0.99 | 0.99 | 1.00 | 1.00 | 1.00 | 1.00 | 0.99 | 0.99 | 0.96 | 0.97 | 0.93 | |
| 2000 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 0.98 | 0.98 | 0.96 | |
| 120 | 1000 | 0.81 | 0.64 | 0.55 | 0.77 | 0.97 | 0.75 | 0.96 | 0.61 | 0.58 | 0.97 | 0.99 | 0.96 |
| 1500 | 1.00 | 0.99 | 0.99 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 0.94 | 0.98 | 0.92 | |
| 2000 | 1.00 | 1.00 | 1.00 | 1.00 | 0.99 | 0.99 | 1.00 | 1.00 | 1.00 | 1.00 | 0.98 | 0.98 |
Table 3. Estimation accuracy of and for Cases 1 and 2 under normal errors. Notes: This table reports the proportions of correct rank selection for Cases 1 and 2 under normal errors. The true global () and local () ranks are set to , yielding four combinations shown in the column groups. Columns headed and report the proportions of correctly selecting the global and local ranks, respectively, while reports the proportion of jointly selecting both ranks correctly. Panels A and B correspond to Cases 1 and 2, respectively.
Additional results under standardized and mixture-normal errors are reported in Section S5.1 of the supplementary material, while Section S5.5 examines sensitivity to the penalty constant. Overall, the simulations indicate generally high rank-selection accuracy, with finite-sample performance depending on the sample dimensions, true ranks, and penalty choice.
Empirical Applications
In this section, we apply the HGFM to returns on S&P 100 constituents from five industry groups. We first estimate the model parameters using the QMLE method. We then compare explanatory fit and investment performance with competing models and examine transaction costs and alternative empirical specifications.
Data Description
We analyze daily returns from the Center for Research in Security Prices (CRSP) for S&P 100 constituents from January 1, 2016, to December 31, 2025. After excluding stocks with insufficient return histories, we obtain a balanced panel of 98 stocks with 2,514 daily observations per stock. The stocks are assigned to five industry groups based on the Global Industry Classification Standard (GICS), with group membership held fixed throughout the sample. Table S12 in the supplementary material reports the group composition and complete stock list. We construct the global and local portfolio weights using market beta, market equity, and book-to-market, each ranked across the sample stocks and scaled to . Local factors use these standardized characteristics within each industry group.
Empirical Findings
Global and Local Factor Model Performance
The information criterion in (14) selects , giving one global factor and one local factor for each industry group. We estimate this specification by QMLE using Algorithm 1 with three random initializations. The global factor is denoted by , and the five local factors by –, following the industry order in Table S12. We compare the full HGFM with the capital asset pricing model (CAPM), the Fama–French three-factor model (FF3) (Fama and French [17]), and the Fama–French five-factor model (FF5) (Fama and French [18]). We also include equal- and value-weighted portfolios of the sample stocks, denoted by EW98 and VW98, respectively. To assess the contributions of the global and local factors, we consider two subsets of the jointly estimated factors: HGFM (global), containing G1, and HGFM (local), containing L1–L5.
Explanatory fit. We assess explanatory performance using Total and cross-sectional (CS ) [30, 24]. Total measures the fit of realized excess returns across assets and dates, whereas CS measures the fit of average excess returns across assets. Specifically,
where , is the th element of , and is the risk-free return. The quantity denotes the fitted excess return from each model. The quantities and denote the time averages of observed and fitted excess returns, respectively.
Table 4 reports Total for the full stock panel and each industry group. We find that the full model has the highest Total among all models considered, both for the full panel and within each industry group. For the full panel, its Total is 0.538, compared with 0.362 for HGFM (global), 0.524 for HGFM (local), and 0.435 for FF5, the best-performing benchmark. The local factors account for much of the explanatory fit, while adding the global factor provides further gains across all five industries. These results indicate that the global and local factors together explain realized return variation more effectively than either subset alone or the benchmark models.
| Model | All | I1 | I2 | I3 | I4 | I5 |
| Panel A: HGFM | ||||||
| HGFM | 0.538 | 0.500 | 0.425 | 0.428 | 0.708 | 0.580 |
| HGFM (global) | 0.362 | 0.323 | 0.305 | 0.241 | 0.540 | 0.362 |
| HGFM (local) | 0.524 | 0.481 | 0.396 | 0.427 | 0.701 | 0.571 |
| Panel B: Benchmarks | ||||||
| CAPM | 0.344 | 0.369 | 0.301 | 0.204 | 0.480 | 0.310 |
| FF3 | 0.425 | 0.413 | 0.347 | 0.222 | 0.634 | 0.427 |
| FF5 | 0.435 | 0.420 | 0.363 | 0.230 | 0.646 | 0.436 |
| EW98 | 0.368 | 0.312 | 0.306 | 0.233 | 0.553 | 0.391 |
| VW98 | 0.332 | 0.374 | 0.295 | 0.207 | 0.451 | 0.284 |
Table 4. Total for the full stock panel and industry groups. Notes: This table reports Total for the full stock panel (All) and the five industry groups (I1–I5). Panel A presents results for the full HGFM and its global-only and local-only subsets, using factors from the joint estimation. Panel B presents results for CAPM, FF3, FF5, and the equal- and value-weighted sample portfolios (EW98 and VW98).
Table 5 reports the corresponding results for CS . For the full panel, HGFM achieves 0.836, exceeding all benchmark models but remaining slightly below HGFM (local). At the industry level, our model outperforms all benchmarks in technology and communication (I1), but not in the other groups. Overall, the benefits of combining global and local factors are particularly evident in explaining realized returns, while improvements in average-return fit vary across industry groups.
| Model | All | I1 | I2 | I3 | I4 | I5 |
| Panel A: HGFM | ||||||
| HGFM | 0.836 | 0.895 | 0.712 | 0.690 | 0.919 | 0.800 |
| HGFM (global) | 0.740 | 0.666 | 0.697 | 0.676 | 0.923 | 0.841 |
| HGFM (local) | 0.839 | 0.851 | 0.798 | 0.692 | 0.916 | 0.849 |
| Panel B: Benchmarks | ||||||
| CAPM | 0.768 | 0.707 | 0.738 | 0.660 | 0.928 | 0.866 |
| FF3 | 0.772 | 0.707 | 0.718 | 0.703 | 0.939 | 0.883 |
| FF5 | 0.767 | 0.707 | 0.683 | 0.712 | 0.937 | 0.889 |
| EW98 | 0.732 | 0.676 | 0.691 | 0.670 | 0.905 | 0.785 |
| VW98 | 0.793 | 0.752 | 0.768 | 0.690 | 0.927 | 0.856 |
Table 5. CS for the full stock panel and industry groups. Notes: This table reports CS for the full stock panel (All) and the five industry groups (I1–I5). The table format follows Table 4.
Investment performance. We evaluate investment performance using mean-variance efficient (MVE) portfolios formed from the factors. For multifactor models, portfolio weights are determined using the means and covariance matrix of daily factor excess returns. Single-factor models use the factor directly. Table 6 reports investment performance before transaction costs. The full HGFM achieves the highest annualized Sharpe ratio among the models considered, at 1.32, compared with 0.84 for HGFM (global), 1.09 for HGFM (local), and 0.98 for FF3. It also has the lowest volatility and maximum drawdown, despite a lower mean return than either factor subset. These results suggest that combining global and local factors improves the balance between return and risk within the evaluation sample.
| Model | AVG (%) | SD (%) | SR | MDD (%) |
| Panel A: HGFM | ||||
| HGFM | 0.75 | 1.48 | 1.32 | 8.12 |
| HGFM (global) | 1.22 | 4.24 | 0.84 | 20.20 |
| HGFM (local) | 1.71 | 4.86 | 1.09 | 28.94 |
| Panel B: Benchmarks | ||||
| CAPM | 1.26 | 4.54 | 0.82 | 24.90 |
| FF3 | 0.84 | 2.32 | 0.98 | 12.23 |
| FF5 | 0.67 | 1.74 | 0.97 | 9.75 |
| EW98 | 1.32 | 4.48 | 0.87 | 21.56 |
| VW98 | 1.38 | 4.43 | 0.93 | 23.64 |
Table 6. Investment performance. Notes: This table reports investment performance across different models. AVG and SD are the mean and standard deviation of monthly total returns (%). SR is the Sharpe ratio of monthly excess returns, annualized by . MDD is the maximum drawdown based on monthly returns (%).
We further examine transaction costs and alternative industry groupings in Tables S13 and S14 of the supplementary material. HGFM retains a higher Sharpe ratio than the equal- and value-weighted sample portfolios at all reported transaction cost rates. The alternative grouping provides an additional check on the specification, showing that explanatory fit and investment performance depend on how industries are grouped.
Economic Interpretation
We interpret the factor portfolios through the estimated characteristic coefficients that determine their stock weights. We report and after the weighted SVD normalization in Proposition 3. The entries of and are the estimated coefficients used to form portfolio weights, rather than coefficients from a separate regression of factors on characteristics. The weights are and before rescaling each portfolio so that its absolute stock weights sum to one. A positive coefficient means that, for two stocks with the same values of the other characteristics, the stock with the higher value of that characteristic receives a higher weight before this rescaling. These coefficients describe portfolio composition, not risk premia.
Table 7 reports the entries of and corresponding to market beta, market equity, and book-to-market. Several patterns emerge from these estimates. First, the global factor assigns greater weight to larger stocks and stocks with higher book-to-market. ratios, while its market-beta coefficient is small and negative. Second, local factors differ across industries. In technology and communication (I1), L1 favors larger, higher-beta stocks with lower book-to-market ratios, whereas L2 in the consumer group (I2) favors larger, lower-beta stocks with higher book-to-market ratios. Third, market equity has the largest coefficient for healthcare (I3), while book-to-market has the largest coefficient for financials (I4) and industrials, energy, and utilities (I5). L5 also assigns greater weight to higher-beta and smaller stocks. These patterns illustrate how observable characteristics shape the global and local portfolio weights. Section S6.4 of the supplementary material reports stock weights and gross-exposure shares for these factor portfolios.
| Factor | Market beta | Market equity | Book-to-market |
| G1 | -0.014 | 0.127 | 0.046 |
| L1 | 0.208 | 0.150 | -0.235 |
| L2 | -0.339 | 0.320 | 0.243 |
| L3 | 0.065 | 0.226 | 0.112 |
| L4 | 0.032 | -0.030 | 0.265 |
| L5 | 0.128 | -0.050 | 0.225 |
Table 7. Estimated characteristic coefficients of global and local factors. Notes: This table reports the estimated coefficients on market beta, market equity, and book-to-market used to construct the global and local portfolio weights. G1 denotes the global factor, and L1–L5 denote the local factors.
Conclusion
This paper proposes a Hierarchical Group Factor Model (HGFM) with global and local factors constructed from observed returns using characteristic-driven weights. Conditional on identification of the time-varying interaction matrix, cross-group characteristic variation separates the global and local coefficient matrices. Normalization using the empirical second-moment matrices of the observed characteristics identifies their factor representations up to paired sign changes. Under a fixed characteristic design and the stated conditions, we establish local consistency for the coefficient and log-variance estimators, as well as consistency and fixed-block asymptotic normality for the normalized factor parameters. We also propose an information criterion that consistently selects the global and common local ranks.
In the application to S&P 100 constituents, the global and local factor portfolios place different weights on stock characteristics. Combining these factors improves realized-return fit across all industry groups relative to either subset alone and the benchmarks, while delivering the highest Sharpe ratio and lowest volatility and maximum drawdown among the portfolios considered. The combined portfolio’s Sharpe ratio remains above those of equal- and value-weighted sample portfolios after transaction costs, although performance varies with industry grouping.
We suggest four directions for future research. First, the model can be extended to incorporate exogenous predictors and autoregressive dynamics, thereby allowing it to capture observed covariate effects and temporal dependence. Second, the framework could be extended to allow the number of characteristics to grow, subject to further identification conditions and asymptotic analysis. Third, this paper focuses on grouped vector-valued observations. Related hierarchical structures could be considered for matrix-valued time series (e.g., [38, 40]), three-level factor structures (e.g., [29]), and tensor factor models (e.g., [11, 10]). Finally, the model could be used to examine global and local components within broader collections of candidate factors ([28]) and incorporate machine-learning-based factor construction.
Data Availability Statement The data that support the findings of this study are available from Wharton Research Data Services. Restrictions apply to the availability of these data, which were used under license for this study.
References
- [1]Ahn, S. C. and A. R. Horenstein (2013). Eigenvalue ratio test for the number of factors. Econometrica 81(3), 1203–1227.DOI
- [2]Ando, T. and J. Bai (2017). Clustering huge number of financial time series: A panel data approach with high-dimensional predictors and factor structures. Journal of the American Statistical Association 112(519), 1182–1198.
- [3]Andreou, E., P. Gagliardini, E. Ghysels, and M. Rubin (2019). Inference in group factor models with an application to mixed-frequency data. Econometrica 87(4), 1267–1305.
- [4]Andreou, E., P. Gagliardini, E. Ghysels, and M. Rubin (2025). Spanning latent and observable factors. Journal of Econometrics 248, 105743.
- [5]Bai, J. (2003). Inferential theory for factor models of large dimensions. Econometrica 71(1), 135–171.DOI
- [6]Bai, J. and K. Li (2012). Statistical analysis of factor models of high dimension. Annals of Statistics 40(1), 436–465.DOI
- [7]Bai, J. and S. Ng (2002). Determining the number of factors in approximate factor models. Econometrica 70(1), 191–221.DOI
- [8]Bai, J. and S. Ng (2013). Principal components estimation and identification of static factors. Journal of Econometrics 176(1), 18–29.DOI
- [9]Brandt, M. W., P. Santa-Clara, and R. Valkanov (2009). Parametric portfolio policies: Exploiting characteristics in the cross-section of equity returns. Review of Financial Studies 22(9), 3411–3447.
- [10]Chen, E. Y., D. Xia, C. Cai, and J. Fan (2024). Semi-parametric tensor factor analysis by iteratively projected singular value decomposition. Journal of the Royal Statistical Society Series B: Statistical Methodology 86(3), 793–823.DOI
- [11]Chen, R., D. Yang, and C.-H. Zhang (2022). Factor models for high-dimensional tensor time series. Journal of the American Statistical Association 117(537), 94–116.
- [12]Choi, I., R. Lin, and Y. Shin (2023). Canonical correlation-based model selection for the multilevel factors. Journal of Econometrics 233(1), 22–44.
- [13]Choi, S. H. and D. Kim (2023). Large volatility matrix analysis using global and national factor models. Journal of Econometrics 235(2), 1917–1933.
- [14]Cong, L. W., G. Feng, J. He, and J. Li (2024). Sparse modeling under grouped heterogeneity with applications to asset pricing. Working paper, https://doi.org/10.2139/ssrn.4511953.
- [15]Cui, L., G. Feng, and J. Yang (2026). Heterogeneous predictability on mutual fund alphas: A sparse clustering GMM approach. Journal of Econometrics 257, 106300.DOI
- [16]Ding, Y., Y. Li, and X. Zheng (2021). High dimensional minimum variance portfolio estimation under statistical factor models. Journal of Econometrics 222(1), 502–515.DOI
- [17]Fama, E. F. and K. R. French (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics 33(1), 3–56.DOI
- [18]Fama, E. F. and K. R. French (2015). A five-factor asset pricing model. Journal of Financial Economics 116(1), 1–22.DOI
- [19]Fan, J., Y. Ke, and Y. Liao (2021). Augmented factor models with applications to validating market risk factors and forecasting bond risk premia. Journal of Econometrics 222(1), 269–294.
- [20]Fan, J. and Y. Liao (2022). Learning latent factors from diversified projections and its applications to over-estimated and weak factors. Journal of the American Statistical Association 117(538), 909–924.
- [21]Fan, J., Y. Liao, and J. Yao (2015). Power enhancement in high dimensional cross- sectional tests. Econometrica 83(4), 1497–1541.arxiv.org/abs/1310.3899
- [22]Fan, J. and H. Peng (2004). Nonconcave penalized likelihood with a diverging number of parameters. Annals of Statistics 32(3), 928–961.DOI
- [23]Feng, G., S. Giglio, and D. Xiu (2020). Taming the factor zoo: A test of new factors. Journal of Finance 75(3), 1327–1370.
- [24]Feng, G., J. He, N. G. Polson, and J. Xu (2024). Deep learning in characteristics-sorted factor models. Journal of Financial and Quantitative Analysis 59(7), 3001–3036.DOI
- [25]Giglio, S. and D. Xiu (2021). Asset pricing with omitted factors. Journal of Political Economy 129(7), 1947–1990.
- [28]Jensen, T. I., B. Kelly, and L. H. Pedersen (2023). Is there a replication crisis in finance? Journal of Finance 78(5), 2465–2518.
- [29]Jin, S., X. Lu, and L. Su (2025). Three-dimensional heterogeneous panel data models with multi-level interactive fixed effects. Journal of Econometrics 249, 105957.
- [30]Kelly, B. T., S. Pruitt, and Y. Su (2019). Characteristics are covariances: A unified model of risk and return. Journal of Financial Economics 134(3), 501–524.
- [31]Lam, C. and Q. Yao (2012). Factor modeling for high-dimensional time series: inference for the number of factors. Annals of Statistics 40(2), 694–726.DOI
- [32]Lee, L.-F. (2004). Asymptotic distributions of quasi-maximum likelihood estimators for spatial autoregressive models. Econometrica 72(6), 1899–1925.DOI
- [33]Lee, L.-f. and J. Yu (2010). Estimation of spatial autoregressive panel data models with fixed effects. Journal of Econometrics 154(2), 165–185.DOI
- [34]Liao, Y. and V. Todorov (2026). Observable versus latent factors. Journal of the American Statistical Association. Forthcoming.DOI
- [35]Ouyang, J., C. Cui, K. M. Tan, and G. Xu (2026). Statistical inference for covariate-adjusted and interpretable generalized latent factor model with application to testing fairness. Annals of Applied Statistics 20(1), 764–788.
- [37]Wan, R., Y. Li, W. Lu, and R. Song (2024). Mining the factor zoo: Estimation of latent factor models with sufficient proxies. Journal of Econometrics 239(2), 105386.arxiv.org/abs/2212.12845
- [38]Wang, D., X. Liu, and R. Chen (2019). Factor models for matrix-valued high-dimensional time series. Journal of Econometrics 208(1), 231–248.arxiv.org/abs/1610.01889
- [39]Wang, F. (2022). Maximum likelihood estimation and inference for high dimensional generalized factor models with application to factor-augmented regressions. Journal of Econometrics 229(1), 180–200.
- [40]Zhang, Y., Y. Hui, J. Song, and S. Zheng (2025). Multilevel matrix factor model. Journal of Econometrics 251, 106033.