|
on Econometrics |
| By: | Soonwoo Kwon; Liyang Sun |
| Abstract: | Researchers often use specifications that correctly estimate the average treatment effect under the assumption of constant effects. When treatment effects are heterogeneous, however, such specifications generally fail to recover this average effect. Augmenting these specifications with interaction terms between demeaned covariates and treatment eliminates this bias, but often leads to imprecise estimates and becomes infeasible under limited overlap. We propose a generalized ridge regression estimator, $\texttt{regulaTE}$, that penalizes the coefficients on the interaction terms to achieve an optimal trade-off between worst-case bias and variance in estimating the average effect under limited treatment effect heterogeneity. Building on this estimator, we construct confidence intervals that remain valid under limited overlap and can also be used to assess sensitivity to violations of the constant effects assumption. We illustrate the method in empirical applications under unconfoundedness and staggered adoption, providing a practical approach to inference under limited overlap. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.05454 |
| By: | Onil Boussim |
| Abstract: | In this paper, we propose a method for correcting sample selection bias when the outcome of interest is categorical, such as occupational choice, health status, or field of study. Classical approaches to sample selection rely on strong parametric distributional assumptions, which may be restrictive in practice. While the recent framework of Chernozhukov et al. (2023) offers a nonparametric identification using a local Gaussian representation (LGR) that holds for any bivariate joint distributions. This makes this approach limited to ordered discrete outcomes. We therefore extend it by developing a local representation that applies to joint probabilities, thereby eliminating the need to impose an artificial ordering on categories. Our representation decomposes each joint probability into marginal probabilities and a category-specific association parameter that captures how selection differentially affects each outcome. Under exclusion restrictions analogous to those in the LGR model, we establish nonparametric point identification of the latent categorical distribution. Building on this identification result, we introduce a semiparametric multinomial logit model with sample selection, propose a computationally tractable two-step estimator, and derive its asymptotic properties. This framework significantly broadens the set of tools available for analyzing selection in categorical and other discrete outcomes, offering substantial relevance for empirical work across economics, health sciences, and social sciences. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.05551 |
| By: | Tatsuru Kikuchi |
| Abstract: | This paper develops a unified theoretical framework for detecting and estimating boundaries in treatment effects across both spatial and temporal dimensions. We formalize the concept of treatment effect boundaries as structural parameters characterizing regime transitions where causal effects cease to operate. Building on diffusion-based models of information propagation, we establish conditions under which spatial and temporal boundaries share common dynamics, derive identification results, and propose consistent estimators. Monte Carlo simulations demonstrate the performance of our methods under various data-generating processes. The framework provides tools for detecting when local treatments become systemic and identifying critical thresholds for policy intervention. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.00754 |
| By: | Zhonghui Zhang (Nanjing Audit University); Chihwa Kao (University of Connecticut); Jungbin Hwang (University of Connecticut) |
| Abstract: | In this paper, we propose a new K-means approach for high-dimensional panel data with unknown group memberships. We highlight that the standard K-means algorithm using Euclidean distance can su¤er from misclassi cation in nite samples due to serial correlation and heteroskedasticity in the panel data. Our proposed weighted K-means algorithm addresses this issue by weighting the Euclidean distance using the full covari-ance structure of idiosyncratic shocks. Assuming that both the cross-sectional and time dimensions of the panel grow large, we develop an asymptotic theory for the weighted K-means algorithm that establishes the consistency of the estimated group centroids and the oracle property for group membership estimation. For practical implemen-tation, we propose a feasible weighted K-means method that employs a regularized estimation of the high-dimensional covariance matrix in the K-means objective func-tion. Monte Carlo simulation results demonstrate the e¤ectiveness of our weighted K-means algorithm in estimating grouped xed-e¤ects models for large panels, partic-ularly when strong serial dependencies exist in both group-level trends and idiosyncratic components. |
| Keywords: | Banding, Grouped xed e¤ects, Heteroscedasticity and autocorrelation, K-means clustering, Sample covariance matrix |
| JEL: | C13 C23 C38 C63 |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:uct:uconnp:2025-09 |
| By: | Matias D. Cattaneo; Michael Jansson; Kenichi Nagasawa |
| Abstract: | This paper develops distribution theory and bootstrap-based inference methods for a broad class of convex pairwise difference estimators. These estimators minimize a kernel-weighted convex-in-parameter function over observation pairs that are similar in terms of certain covariates, where the similarity is governed by a localization (bandwidth) parameter. While classical results establish asymptotic normality under restrictive bandwidth conditions, we show that valid Gaussian and bootstrap-based inference remains possible under substantially weaker assumptions. First, we extend the theory of small bandwidth asymptotics to convex pairwise estimation settings, deriving robust Gaussian approximations even when a smaller than standard bandwidth is used. Second, we employ a debiasing procedure based on generalized jackknifing to enable inference with larger bandwidths, while preserving convexity of the objective function. Third, we construct a novel bootstrap method that adjusts for bandwidth-induced variance distortions, yielding valid inference across a wide range of bandwidth choices. Our proposed inference method enjoys demonstrable more robustness, while retaining the practical appeal of convex pairwise difference estimators. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.05991 |
| By: | Sid Kankanala |
| Abstract: | This paper develops a generalized (quasi-) Bayes framework for conditional moment restriction models, where the parameter of interest is a nonparametric structural function of endogenous variables. We establish contraction rates for a class of Gaussian process priors and provide conditions under which a Bernstein-von Mises theorem holds for the quasi-Bayes posterior. Consequently, we show that optimally weighted quasi-Bayes credible sets achieve exact asymptotic frequentist coverage, extending classical results for parametric GMM models. As an application, we estimate firm-level production functions using Chilean plant-level data. Simulations illustrate the favorable performance of generalized Bayes estimators relative to common alternatives. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.01036 |
| By: | Peiyun Jiang; Takashi Yamagata |
| Abstract: | In this paper, we propose a novel bootstrap algorithm that is more efficient than existing methods for approximating the distribution of the factor-augmented regression estimator for a rotated parameter vector. The regression is augmented by $r$ factors extracted from a large panel of $N$ variables observed over $T$ time periods. We consider general weak factor (WF) models with $r$ signal eigenvalues that may diverge at different rates, $N^{\alpha _{k}}$, where $0 |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.00947 |
| By: | G\"ozde Sert; Abhishek Chakrabortty; Anirban Bhattacharya |
| Abstract: | Inference in semi-supervised (SS) settings has gained substantial attention in recent years due to increased relevance in modern big-data problems. In a typical SS setting, there is a much larger-sized unlabeled data, containing only observations of predictors, and a moderately sized labeled data containing observations for both an outcome and the set of predictors. Such data naturally arises when the outcome, unlike the predictors, is costly or difficult to obtain. One of the primary statistical objectives in SS settings is to explore whether parameter estimation can be improved by exploiting the unlabeled data. We propose a novel Bayesian method for estimating the population mean in SS settings. The approach yields estimators that are both efficient and optimal for estimation and inference. The method itself has several interesting artifacts. The central idea behind the method is to model certain summary statistics of the data in a targeted manner, rather than the entire raw data itself, along with a novel Bayesian notion of debiasing. Specifying appropriate summary statistics crucially relies on a debiased representation of the population mean that incorporates unlabeled data through a flexible nuisance function while also learning its estimation bias. Combined with careful usage of sample splitting, this debiasing approach mitigates the effect of bias due to slow rates or misspecification of the nuisance parameter from the posterior of the final parameter of interest, ensuring its robustness and efficiency. Concrete theoretical results, via Bernstein--von Mises theorems, are established, validating all claims, and are further supported through extensive numerical studies. To our knowledge, this is possibly the first work on Bayesian inference in SS settings, and its central ideas also apply more broadly to other Bayesian semi-parametric inference problems. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.17385 |
| By: | David Arbour; Harsh Parikh; Bijan Niknam; Elizabeth Stuart; Kara Rudolph; Avi Feller |
| Abstract: | Many common estimators in machine learning and causal inference are linear smoothers, where the prediction is a weighted average of the training outcomes. Some estimators, such as ordinary least squares and kernel ridge regression, allow for arbitrarily negative weights, which improve feature imbalance but often at the cost of increased dependence on parametric modeling assumptions and higher variance. By contrast, estimators like importance weighting and random forests (sometimes implicitly) restrict weights to be non-negative, reducing dependence on parametric modeling and variance at the cost of worse imbalance. In this paper, we propose a unified framework that directly penalizes the level of extrapolation, replacing the current practice of a hard non-negativity constraint with a soft constraint and corresponding hyperparameter. We derive a worst-case extrapolation error bound and introduce a novel "bias-bias-variance" tradeoff, encompassing biases due to feature imbalance, model misspecification, and estimator variance; this tradeoff is especially pronounced in high dimensions, particularly when positivity is poor. We then develop an optimization procedure that regularizes this bound while minimizing imbalance and outline how to use this approach as a sensitivity analysis for dependence on parametric modeling assumptions. We demonstrate the effectiveness of our approach through synthetic experiments and a real-world application, involving the generalization of randomized controlled trial estimates to a target population of interest. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.17180 |
| By: | Karsten Reichold; Ulrike Schneider |
| Abstract: | This paper establishes new asymptotic results for the adaptive LASSO estimator in cointegrating regression models. We study model selection probabilities, estimator consistency, and limiting distributions under both standard and moving-parameter asymptotics. We also derive uniform convergence rates and the fastest local-to-zero rates that can still be detected by the estimator, complementing and extending the results of Lee, Shi, and Gao (2022, Journal of Econometrics, 229, 322--349). Our main findings include that under conservative tuning, the adaptive LASSO estimator is uniformly $T$-consistent and the cut-off rate for local-to-zero coefficients that can be detected by the procedure is $1/T$. Under consistent tuning, however, both rates are slower and depend on the tuning parameter. The theoretical results are complemented by a detailed simulation study showing that the finite-sample distribution of the adaptive LASSO estimator deviates substantially from what is suggested by the oracle property, whereas the limiting distributions derived under moving-parameter asymptotics provide much more accurate approximations. Finally, we show that our results also extend to models with local-to-unit-root regressors and to predictive regressions with unit-root predictors. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.07204 |
| By: | Ziyan Zhao (Economic Growth Centre, School of Social Sciences, Nanyang Technological University); Qingfeng Liu (Department of Industrial and Systems Engineering, Hosei University) |
| Abstract: | This study proposes a time-varying structural approximate dynamic factor (TVS-ADF) model by extending the ADF model in state-space form. The TVS-ADF model considers time-varying coefficients and a time-varying variance–covariance matrix of its innovation terms, so that it can capture complex dynamic economic characteris- tics. We propose the identification scheme of the common factors in the TVS-ADF and derive the identification theory. We also propose an effective Markov chain Monte Carlo (MCMC) algorithm to estimate the TVS-ADF. To avoid the overparameterization caused by the time-varying characteristics of the TVS-ADF, we include the shrinkage and sparsification approaches in the MCMC algorithm. Additionally, we propose several effective information criteria for the determination of the number of factors in the TVS-ADF. Extensive artificial simulations demonstrate that the TVS-ADF has better forecast performance than the ADF in almost all settings for different numbers of explained variables, numbers of explanatory variables, sparsity levels, and sample sizes. An empirical application to macroeconomic forecasting also indicates that our model can substantially improve predictive accuracy and capture the dynamic features of an economic system better than the ADF. |
| Keywords: | MCMC, shrinkage, sparsification, overparameterization, algorithms |
| Date: | 2024–01 |
| URL: | https://d.repec.org/n?u=RePEc:nan:wpaper:2401 |
| By: | Jungbin Hwang (University of Connecticut); Feifan Wang (University of Connecticut) |
| Abstract: | Robust inference in bond predictive regressions faces challenges due to strong time-series per-sistence and unknown cross-sectional factor structures in the bond yield vector. These diffi-culties are particularly pronounced in analyzing the spanning hypothesis, which tests whether factors beyond the first three principal components (PCs)—level, slope, and curvature—improve bond return predictability. To address this, we develop a novel nonparametric sieve bootstrap approach for multivariate bond yield data with different maturities. Our method provides accurate size and improved power performance in bond predictive regression, com-pared to existing bootstrap inference procedures for the spanning hypothesis. Revisiting Cochrane and Piazzesi (2005)’s return-forecasting factor, we find strong evidence of its pre-dictive power beyond the first three PCs for bond excess returns in most sample periods after the 1960s. However, we find that these predictive gains significantly decline when the sample period extends to include recent years after 2019. |
| Keywords: | Sieve bootstrap, Term structure of interest rates, Predictive regression, Spanning hypothesis |
| JEL: | C12 C15 E43 G12 |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:uct:uconnp:2025-10 |
| By: | Antonio Cozzolino (NYU Stern, New York University, New York, NY, USA); Cristina Gualdani (School of Economics and Finance, Queen Mary University of London, London, UK); Ivan Gufler (Department of Economics and Finance, University of Bonn, Bonn, Germany); Niccolò Lomys (CSEF and Department of Economics and Statistics, University of Naples Federico II, Naples, Italy); Lorenzo Magnolfi (Department of Economics, University of Wisconsin-Madison, Madison, WI, USA) |
| Abstract: | We develop an econometric framework for recovering structural primitives---such as marginal costs---from price or quantity data generated by firms whose decisions are governed by reinforcement-learning algorithms. Guided by recent theory and simulations showing that such algorithms can learn to approximate repeated-game equilibria, we impose only the minimal optimality conditions implied by equilibrium, while remaining agnostic about the algorithms’ hidden design choices and the resulting conduct---competitive, collusive, or anywhere in between. These weak restrictions yield set identification of the primitives; we characterise the resulting sets and construct estimators with valid confidence regions. Monte~Carlo simulations confirm that our bounds contain the true parameters across a wide range of algorithm specifications, and that the sets tighten substantially when exogenous demand variation across markets is exploited. The framework thus offers a practical tool for empirical analysis and regulatory assessment of algorithmic behaviour. |
| Keywords: | Algorithms; Reinforcement Learning; Repeated Games; Coarse Correlated Equilibrium; Partial Identification; Incomplete Models |
| JEL: | C1 C5 C7 D8 L1 |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:net:wpaper:2504 |
| By: | Bowen Fu; Chenghan Hou; Jan Pr\"user |
| Abstract: | This paper proposes a structural multivariate unobserved components model with external instrument (SMUC-IV) to investigate the effects of monetary policy shocks on key U.S. macroeconomic "stars"-namely, the level of potential output, the growth rate of potential output, trend inflation, and the neutral interest rate. A key feature of our approach is the use of an external instrument to identify monetary policy shocks within the multivariate unob- served components modeling framework. We develop an MCMC estimation method to facilitate posterior inference within our proposed SMUC-IV frame- work. In addition, we propose an marginal likelihood estimator to enable model comparison across alternative specifications. Our empirical analysis shows that contractionary monetary policy shocks have significant negative effects on the macroeconomic stars, highlighting the nonzero long-run effects of transitory monetary policy shocks. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.05802 |
| By: | Leonardo N. Ferreira; Haroon Mumtaz; Ana Skoblar |
| Abstract: | This paper introduces a Bayesian vector autoregression (BVAR) with stochastic volatility-in-mean and time-varying skewness. Unlike previous approaches, the proposed model allows both volatility and skewness to directly affect macroeconomic variables. We provide a Gibbs sampling algorithm for posterior inference and apply the model to quarterly data for the US and the UK. Empirical results show that skewness shocks have economically significant effects on output, inflation and spreads, often exceeding the impact of volatility shocks. In a pseudo-real-time forecasting exercise, the proposed model outperforms existing alternatives in many cases. Moreover, the model produces sharper measures of tail risk, revealing that standard stochastic volatility models tend to overstate uncertainty. These findings highlight the importance of incorporating time-varying skewness for capturing macro-financial risks and improving forecast performance. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.08415 |
| By: | Thomas T. Yang |
| Abstract: | We revisit tail-index regressions. For linear specifications, we find that the usual full-rank condition can fail because conditioning on extreme outcomes causes regressors to degenerate to constants. More generally, the conditional distribution of the covariates in the tails concentrates on the values at which the tail index is minimized. Away from those points, the conditional density tends to zero. For local nonparametric tail index regression, the convergence rate can be very slow. We conclude with practical suggestions for applied work. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.01535 |
| By: | Maria Gadea; Òscar Jordà |
| Abstract: | Bootstrap procedures for local projections typically rely on assuming that the data generating process (DGP) is a finite order vector autoregression (VAR), often taken to be that implied by the local projection at horizon 1. Although convenient, it is well documented that a VAR can be a poor approximation to impulse dynamics at horizons beyond its lag length. In this paper we assume instead that the precise form of the parametric model generating the data is not known. If one is willing to assume that the DGP is perhaps an infinite order process, a larger class of models can be accommodated and more tailored bootstrap procedures can be constructed. Using the moving average representation of the data, we construct appropriate bootstrap procedures. |
| Keywords: | local projections; inference |
| JEL: | C31 C32 |
| Date: | 2025–09–25 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedfwp:101873 |
| By: | Nicolas Salvad\'e; Tim Hillel |
| Abstract: | In this paper, we present a general specification for Functional Effects Models, which use Machine Learning (ML) methodologies to learn individual-specific preference parameters from socio-demographic characteristics, therefore accounting for inter-individual heterogeneity in panel choice data. We identify three specific advantages of the Functional Effects Model over traditional fixed, and random/mixed effects models: (i) by mapping individual-specific effects as a function of socio-demographic variables, we can account for these effects when forecasting choices of previously unobserved individuals (ii) the (approximate) maximum-likelihood estimation of functional effects avoids the incidental parameters problem of the fixed effects model, even when the number of observed choices per individual is small; and (iii) we do not rely on the strong distributional assumptions of the random effects model, which may not match reality. We learn functional intercept and functional slopes with powerful non-linear machine learning regressors for tabular data, namely gradient boosting decision trees and deep neural networks. We validate our proposed methodology on a synthetic experiment and three real-world panel case studies, demonstrating that the Functional Effects Model: (i) can identify the true values of individual-specific effects when the data generation process is known; (ii) outperforms both state-of-the-art ML choice modelling techniques that omit individual heterogeneity in terms of predictive performance, as well as traditional static panel choice models in terms of learning inter-individual heterogeneity. The results indicate that the FI-RUMBoost model, which combines the individual-specific constants of the Functional Effects Model with the complex, non-linear utilities of RUMBoost, performs marginally best on large-scale revealed preference panel data. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.18047 |
| By: | Lamia Lamrani; Beno\^it Collins; Jean-Philippe Bouchaud |
| Abstract: | Cross-validation is one of the most widely used methods for model selection and evaluation; its efficiency for large covariance matrix estimation appears robust in practice, but little is known about the theoretical behavior of its error. In this paper, we derive the expected Frobenius error of the holdout method, a particular cross-validation procedure that involves a single train and test split, for a generic rotationally invariant multiplicative noise model, therefore extending previous results to non-Gaussian data distributions. Our approach involves using the Weingarten calculus and the Ledoit-P\'ech\'e formula to derive the oracle eigenvalues in the high-dimensional limit. When the population covariance matrix follows an inverse Wishart distribution, we approximate the expected holdout error, first with a linear shrinkage, then with a quadratic shrinkage to approximate the oracle eigenvalues. Under the linear approximation, we find that the optimal train-test split ratio is proportional to the square root of the matrix dimension. Then we compute Monte Carlo simulations of the holdout error for different distributions of the norm of the noise, such as the Gaussian, Student, and Laplace distributions and observe that the quadratic approximation yields a substantial improvement, especially around the optimal train-test split ratio. We also observe that a higher fourth-order moment of the Euclidean norm of the noise vector sharpens the holdout error curve near the optimal split and lowers the ideal train-test ratio, making the choice of the train-test ratio more important when performing the holdout method. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.13923 |
| By: | Partha Deb; Edward C. Norton; Jeffrey M. Wooldridge; Jeffrey E. Zabel |
| Abstract: | For difference-in-differences methods, there has been great attention to obtaining consistent estimates of treatment effects, especially when the treatment effects are heterogeneous. However, there has been little discussion of the importance of weights in aggregating those treatment effects into an overall average treatment effect on the treated. There are many possible ways to aggregate estimated cohort-time treatment effects. We show that the standard software used to estimate Callaway and Sant’Anna’s method uses weights that are not just the number of treated observations in treated years. Instead, the software uses weights that include the number of observations in the reference pre-period instead of only the number of observations in the treated periods. We discuss why the aggregation weights matter and under what circumstances the weights make the most difference. |
| JEL: | C10 C18 |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34331 |
| By: | Reca Sarfati; Vod Vilfort |
| Abstract: | Pre-analysis plans (PAPs) have become standard in experimental economics research, but it is nevertheless common to see researchers deviating from their PAPs to supplement preregistered estimates with non-prespecified findings. While such ex-post analysis can yield valuable insights, there is broad uncertainty over how to interpret -- or whether to even acknowledge -- non-preregistered results. In this paper, we consider the case of a truth-seeking researcher who, after seeing the data, earnestly wishes to report additional estimates alongside those preregistered in their PAP. We show that, even absent "nefarious" behavior, conventional confidence intervals and point estimators are invalid due to the fact that non-preregistered estimates are only reported in a subset of potential data realizations. We propose inference procedures that account for this conditional reporting. We apply these procedures to Bessone et al. (2021), which studies the economic effects of increased sleep among the urban poor. We demonstrate that, depending on the reason for deviating, the adjustments from our procedures can range from having no difference to an economically significant difference relative to conventional practice. Finally, we consider the robustness of our procedure to certain forms of misspecification, motivating possible heuristic checks and norms for journals to adopt. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.02507 |
| By: | Hrishikesh Vinod (Fordham University) |
| Abstract: | Statistical measure(s) of dependence (MOD) between variables are essential for most empirical work. We show that Renyi’s postulates from the 1950s are responsible for serious MOD limitations. (i) They rule out examples when one of the variables is deterministic (like time or age), (ii) They are always positive, implying no one-tailed significance tests. (iii) They disallow ubiquitous asymmetric MOD. Many MOD exist in the literature, including those from 2022 and 2025, share these limitations because they fail to satisfy our three new axioms. We also describe a new implementation of a powerful one-sided test for the null of zero Pearson correlation with Taraldsen’s exact sampling distribution and provide a new table for practitioners. We include a published example where Taraldsen’s test makes a practical difference. The code to implement all our proposals is in R packages. |
| Keywords: | Kernel Regression, Generalized Correlation, Asymmetric Dependence, Exact t-density |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:frd:wpaper:dp2025-02er:dp2025-02 |
| By: | Natascha Hey; Eyal Neuman; Sturmius Tuschmann |
| Abstract: | We introduce an offline nonparametric estimator for concave multi-asset propagator models based on a dataset of correlated price trajectories and metaorders. Compared to parametric models, our framework avoids parameter explosion in the multi-asset case and yields confidence bounds for the estimator. We implement the estimator using both proprietary metaorder data from Capital Fund Management (CFM) and publicly available S&P order flow data, where we augment the former dataset using a metaorder proxy. In particular, we provide unbiased evidence that self-impact is concave and exhibits a shifted power-law decay, and show that the metaorder proxy stabilizes the calibration. Moreover, we find that introducing cross-impact provides a significant gain in explanatory power, with concave specifications outperforming linear ones, suggesting that the square-root law extends to cross-impact. We also measure asymmetric cross-impact between assets driven by relative liquidity differences. Finally, we demonstrate that a shape-constrained projection of the nonparametric kernel not only ensures interpretability but also slightly outperforms established parametric models in terms of predictive accuracy. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.06879 |
| By: | Tang, Bo Rui; Zhu, Jin; Wang, Ting Yin; Zhu, Junxian |
| Abstract: | In this paper, we examine the problem of sliced inverse regression (SIR), a widely used method for sufficient dimension reduction (SDR). It was designed to find reduced-dimensional versions of multivariate predictors by replacing them with a minimally adequate collection of their linear combinations without loss of information. Recently, regularization methods have been proposed in SIR to incorporate a sparse structure of predictors for better interpretability. However, existing methods consider convex relaxation to bypass the sparsity constraint, which may not lead to the best subset, and particularly tends to include irrelevant variables when predictors are correlated. In this paper, we approach sparse SIR as a nonconvex optimization problem and directly tackle the sparsity constraint by establishing the optimal conditions and iteratively solving them via the splicing technique. Without employing convex relaxation on the sparsity constraint and the orthogonal constraint, our algorithm exhibits superior empirical merits, as evidenced by extensive numerical studies. Computationally, our algorithm is much faster than the relaxed approach for the natural sparse SIR estimator. Statistically, our algorithm surpasses existing methods in terms of accuracy for central subspace estimation and best subset selection and sustains high performance even with correlated predictors. |
| Keywords: | best subset selection; nonconvex optimization; optimal conditions; sliced inverse regression; sparsity constraint; splicing technique |
| JEL: | C1 |
| Date: | 2025–05–31 |
| URL: | https://d.repec.org/n?u=RePEc:ehl:lserod:129705 |
| By: | Emerson Melo; David M\"uller |
| Abstract: | We establish a link between a class of discrete choice models and the theory of online learning and multi-armed bandits. Our contributions are: (i) sublinear regret bounds for a broad algorithmic family, encompassing Exp3 as a special case; (ii) a new class of adversarial bandit algorithms derived from generalized nested logit models \citep{wen:2001}; and (iii) \textcolor{black}{we introduce a novel class of generalized gradient bandit algorithms that extends beyond the widely used softmax formulation. By relaxing the restrictive independence assumptions inherent in softmax, our framework accommodates correlated learning dynamics across actions, thereby broadening the applicability of gradient bandit methods.} Overall, the proposed algorithms combine flexible model specification with computational efficiency via closed-form sampling probabilities. Numerical experiments in stochastic bandit settings demonstrate their practical effectiveness. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.03979 |
| By: | Samuel N. Cohen; Cephas Svosve |
| Abstract: | We explore a link between stochastic volatility (SV) and path-dependent volatility (PDV) models. Using assumed density filtering, we map a given SV model into a corresponding PDV representation. The resulting specification is lightweight, improves in-sample fit, and delivers robust out-of-sample forecasts. We also introduce a calibration procedure for both SV and PDV models that produces standard errors for parameter estimates and supports joint calibration of SPX/VIX smile. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.02024 |
| By: | Miguel Alves Pereira |
| Abstract: | This article proposes predictive economics as a distinct analytical perspective within economics, grounded in machine learning and centred on predictive accuracy rather than causal identification. Drawing on the instrumentalist tradition (Friedman), the explanation-prediction divide (Shmueli), and the contrast between modelling cultures (Breiman), we formalise prediction as a valid epistemological and methodological objective. Reviewing recent applications across economic subfields, we show how predictive models contribute to empirical analysis, particularly in complex or data-rich contexts. This perspective complements existing approaches and supports a more pluralistic methodology - one that values out-of-sample performance alongside interpretability and theoretical structure. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.04726 |
| By: | Axel Ciceri; Austin Cottrell; Joshua Freeland; Daniel Fry; Hirotoshi Hirai; Philip Intallura; Hwajung Kang; Chee-Kong Lee; Abhijit Mitra; Kentaro Ohno; Das Pemmaraju; Manuel Proissl; Brian Quanz; Del Rajan; Noriaki Shimada; Kavitha Yograj |
| Abstract: | The estimation of fill probabilities for trade orders represents a key ingredient in the optimization of algorithmic trading strategies. It is bound by the complex dynamics of financial markets with inherent uncertainties, and the limitations of models aiming to learn from multivariate financial time series that often exhibit stochastic properties with hidden temporal patterns. In this paper, we focus on algorithmic responses to trade inquiries in the corporate bond market and investigate fill probability estimation errors of common machine learning models when given real production-scale intraday trade event data, transformed by a quantum algorithm running on IBM Heron processors, as well as on noiseless quantum simulators for comparison. We introduce a framework to embed these quantum-generated data transforms as a decoupled offline component that can be selectively queried by models in low-latency institutional trade optimization settings. A trade execution backtesting method is employed to evaluate the fill prediction performance of these models in relation to their input data. We observe a relative gain of up to ~ 34% in out-of-sample test scores for those models with access to quantum hardware-transformed data over those using the original trading data or transforms by noiseless quantum simulation. These empirical results suggest that the inherent noise in current quantum hardware contributes to this effect and motivates further studies. Our work demonstrates the emerging potential of quantum computing as a complementary explorative tool in quantitative finance and encourages applied industry research towards practical applications in trading. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.17715 |
| By: | Tonghui Qi |
| Abstract: | Many auction datasets with reserve prices do not include bids that fall below the reserve. This paper establishes nonparametric identification results in first- and second-price auctions when transaction prices are truncated by a binding reserve price under a range of information structures. In the simplest case-where the number of potential bidders is fixed and known across all auctions-if only the transaction price is observed, the bidders' private-value distribution is identified in second-price auctions but not in first-price auctions. Identification in first-price auctions can be achieved if either the number of active bidders (those whose bids exceed the reserve) or the number of auctions with no sales (all bids below the reserve) is observed. When the number of potential bidders varies across auctions and is unknown, the bidders' private-value distribution is identified in first-price auctions but not in second-price auctions, provided that both the transaction price and the number of active bidders are observed. Finally, I extend these results to auctions with entry costs, which face a similar truncation issue when data on potential bidders who do not enter are missing. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.04464 |
| By: | Martin Aichele; Igor Cialenco; Damian Jelito; Marcin Pitera |
| Abstract: | We develop a statistical framework for risk estimation, inspired by the axiomatic theory of risk measures. Coherent risk estimators -- functionals of P&L samples inheriting the economic properties of risk measures -- are defined and characterized through robust representations linked to $L$-estimators. The framework provides a canonical methodology for constructing estimators with sound financial and statistical properties, unifying risk measure theory, principles for capital adequacy, and practical statistical challenges in market risk. A numerical study illustrates the approach, focusing on expected shortfall estimation under both i.i.d. and overlapping samples relevant for regulatory FRTB model applications. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.05809 |