arXiv subjects

Charles Shaw

Publications and source records attributed to Charles Shaw.

8 recordsLinked to original sources

arXiv2025

srvar-toolkit: A Python Implementation of Shadow-Rate Vector Autoregressions with Stochastic Volatility

We introduce srvar-toolkit, an open-source Python package for Bayesian vector autoregression with shadow-rate constraints and stochastic volatility. The toolkit implements the methodology of Grammatikopoulos (2025, Journal of Forecasting) for forecasting macroeconomic variables when interest rates hit the effective lower bound. We provide conjugate Normal-Inverse-Wishart priors with Minnesota-style shrinkage, latent shadow-rate data augmentation via Gibbs sampling, diagonal stochastic volatility using the Kim-Shephard-Chib mixture approximation, and stochastic search variable selection. Core dependencies are NumPy, SciPy, and Pandas, with optional extras for plotting and a configuration-driven command-line interface. We release the software under the MIT licence at https://github.com/shawcharles/srvar-toolkit.

stat.CO↗

arXiv2025

A Path Signature Framework for Detecting Creative Fatigue in Digital Advertising

This paper introduces a signature-based framework for detecting advertising creative fatigue using path signatures, a geometric representation from rough path theory. Creative fatigue -- the degradation of creative effectiveness under repeated exposure -- is operationally important in digital marketing because delayed detection can translate directly into avoidable opportunity cost. We reframe fatigue monitoring as a geometric change detection problem: advertising performance trajectories are embedded as paths and represented by truncated (log-)signatures, enabling detection of changes in trend, volatility, and non-linear dynamics beyond simple mean or variance shifts. We further connect statistical detection to managerial decision-making via an explicit quantification of performance loss relative to a benchmark period. Because proprietary production data cannot be released, we evaluate the proposed framework on a synthetic panel dataset designed to mimic realistic impression volumes and noisy day-to-day CTR dynamics. We define observed CTR as the realised binomial rate $CTR_t := C_t/I_t$ using daily clicks $C_t$ and impressions $I_t$. The accompanying CSV also contains a pre-computed CTR field (e.g., due to rounding or upstream derivation), but all modelling and evaluation in this paper use $C_t/I_t$. Crucially, the dataset does not include injected changepoints; we therefore define an operational ground truth for ``fatigue onset'' based on a noise-robust CTR estimate and a sustained deterioration relative to a recent-best baseline. We report lead-time (early warning) and alert-burden metrics under this operational definition, and provide a sensitivity analysis over the detector's primary tuning parameters. The methodology scales linearly in time-series length for fixed signature depth and is suitable for monitoring large creative portfolios.

stat.AP↗

arXiv2025

Optimized Supergeo Design: A Scalable Framework for Geographic Marketing Experiments

Geographic experiments are a widely-used methodology for measuring incremental return on ad spend (iROAS) at scale, yet their design presents significant challenges. The unit count is small, heterogeneity is large, and the optimal Supergeo partitioning problem is NP-hard. We introduce Optimized Supergeo Design (OSD), a two-stage framework that renders Supergeo designs practical for large-scale markets. Principal Component Analysis (PCA) first reduces the covariate space to create interpretable geo-embeddings. A Mixed-Integer Linear Programming (MILP) solver then selects a partition that balances both baseline outcomes and pre-treatment covariates. We provide theoretical arguments that OSD's objective value is within $(1+\varepsilon)$ of the global optimum under community-structure assumptions. Rigorous ablation analysis on synthetic data shows that PCA- and random-embedding Supergeo designs match unit-level randomisation in estimation error while delivering tighter covariate balance, whereas spectral embeddings substantially worsen both RMSE and balance. Crucially, OSD solves the scalability bottleneck. For $N=210$ markets, OSD completes in a fraction of a second, while exact Supergeo covering MIPs described in prior work are projected to require orders of magnitude longer, on the order of weeks. Scalability experiments up to $N=1\,000$ units show that OSD remains fast without trimming markets. In our main synthetic setting with $N=200$ units, PCA- and random-embedding designs keep covariate imbalance at only a few percentage points while preserving every media dollar, establishing a scalable framework that matches the statistical efficiency of randomisation with the operational practicality of Supergeos.

stat.AP↗

arXiv2025

Model Risk in Machine-Learning Distributional IV Estimation

We study model risk in machine-learning estimation of the Distributional Instrumental Variable Local Average Treatment Effect (D-IV-LATE), the distributional IV effect for the subpopulation induced into treatment by the instrument. The contribution is not a new neural causal estimand. We implement a reduced-form orthogonal level-score DML estimator for the covariate-adjusted D-IV-LATE target and use it to ask how much the nuisance learner matters for distributional IV conclusions. In simulations with explicit monotone principal strata and known complier truth, Kolmogorov-Arnold Networks (KANs) are faster than Random Forests in every scenario examined, but Random Forests usually produce more accurate D-IV-LATE curves. A targeted KAN ablation selects a width-64 KAN as the best KAN variant among those tested, but this is a speed/accuracy tradeoff rather than evidence of KAN dominance. In a 401(k) application, RF and KAN estimates differ materially, with frequent sign reversals along the estimated curve. The KAN instrument-propensity estimates also concentrate near the boundaries, so the KAN empirical curve is best read as sensitivity evidence. In inference validation, KAN pointwise intervals undercover badly under both asymptotic and bootstrap constructions, while RF asymptotic intervals are better calibrated in the validation designs. The main lesson is cautionary and constructive: speed and architectural flexibility are not enough for causal inference. Applied researchers using ML-based distributional IV estimators would be advised to benchmark nuisance learners, report overlap and calibration diagnostics, and validate inference directly.

econ.EM↗

arXiv2024

Financial Performance and Innovation: Evidence From USA, 1998-2023

This study explores the relationship between R&D intensity, as a measure of innovation, and financial performance among S&P 500 companies over 100 quarters from 1998 to 2023, including multiple crisis periods. It challenges the conventional wisdom that larger companies are more prone to innovate, using a comprehensive dataset across various industries. The analysis reveals diverse associations between innovation and key financial indicators such as firm size, assets, EBITDA, and tangibility. Our findings underscore the importance of innovation in enhancing firm competitiveness and market positioning, highlighting the effectiveness of countercyclical innovation policies. This research contributes to the debate on the role of R&D investments in driving firm value, offering new insights for both academic and policy discussions.

econ.GN↗

arXiv2022

Utilitarianism on the front lines: COVID-19, public ethics, and the "hidden assumption" problem

How should we think of the preferences of citizens? Whereas self-optimal policy is relatively straightforward to produce, socially optimal policy often requires a more detailed examination. In this paper, we identify an issue that has received far too little attention in welfarist modelling of public policy, which we name the "hidden assumptions" problem. Hidden assumptions can be deceptive because they are not expressed explicitly and the social planner (e.g. a policy maker, a regulator, a legislator) may not give them the critical attention they need. We argue that ethical expertise has a direct role to play in public discourse because it is hard to adopt a position on major issues like public health policy or healthcare prioritisation without making contentious assumptions about population ethics. We then postulate that ethicists are best situated to critically evaluate these hidden assumptions, and can therefore play a vital role in public policy debates.

econ.GN↗

arXiv2022

Portfolio Diversification Revisited

We relax a number of assumptions in Alexeev and Tapon (2012) in order to account for non-normally distributed, skewed, multi-regime, and leptokurtic asset return distributions. We calibrate a Markov-modulated Levy process model to equity market data to demonstrate the merits of our approach, and show that the calibrated models do a good job of matching the empirical moments. Finally, we argue that much of the related literature on portfolio diversification relies on assumptions that are in tension with certain observable regularities and which, if ignored, may lead to underestimation of risk.

q-fin.PM↗

arXiv2018

Conditional heteroskedasticity in crypto-asset returns

This paper examines the time series properties of cryptocurrency assets, such as Bitcoin, using established econometric inference techniques, namely models of the GARCH family. The contribution of this study is twofold. I explore the time series properties of cryptocurrencies, a new type of financial asset on which there appears to be little or no literature. I suggest an improved econometric specification to that which has been recently proposed in Chu et al (2017), the first econometric study to examine the price dynamics of the most popular cryptocurrencies. Questions regarding the reliability of their study stem from the authors mis-diagnosing the distribution of GARCH innovations. Checks are performed on whether innovations are Gaussian or GED by using Kolmogorov type non-parametric tests and Khmaladze's martingale transformation. Null of gaussianity is strongly rejected for all GARCH(p,q) models, with $p,q \in \{1,\ldots,5 \}$, for all cryptocurrencies in sample. For tests of normality, I make use of the Gauss-Kronrod quadrature. Parameters of GARCH models are estimated with generalized error distribution innovations using maximum likelihood. For calculating P-values, the parametric bootstrap method is used. Arguing against Chu et al (2017), I show that there is a strong empirical argument against modelling innovations under some common assumptions.

q-fin.ST↗