Searcharxiv⌕ Search

arXiv subjects

Adam Butler

Publications and source records attributed to Adam Butler.

6 recordsLinked to original sources

Idiosyncratic Market Effects in Global Time Series Momentum

Time series momentum is a pervasive explanatory factor in financial markets whereby a market's recent historical average return positively predicts future returns. Recent applications of statistical learning augment the traditional linear model definition with a more descriptive global nonlinear factor model. We extend the work of \citet{MoskowitzNonlinear} and demonstrate that models fit at the individual market level exhibit response curves that are distinct from the global factor model. We highlight the heterogeneity of forward market returns as a function of time series momentum and provide a framework that enables each feature-market pair to express its unique prediction behaviour. Empirical results demonstrate that, despite substantially smaller sample sizes, fitting nonlinear models to individual markets is a robust and often overlooked modelling approach that can result in a marginal improvement in out-of-sample performance.

q-fin.MF↗

Global Seasonality Factor Investing

We study seasonality as a factor for predicting risk-adjusted returns across global futures markets. We document five seasonality anomalies that, in aggregate, are pervasive across bond, commodity and equity markets. We demonstrate that each seasonality factor produces excess returns that are not reasonably explained by established global time series carry, momentum and skewness factors. We apply recent innovations in statistical learning and fit nonlinear seasonality models at the individual market level, thereby allowing each feature-market pair to express its unique seasonal prediction pattern. We discuss candidate economic rationales for the existence and persistence of each seasonality anomaly. Walk-forward out-of-sample testing demonstrates that fast-moving seasonal patterns, such as day-of-week and day-of-month, produce strong gross performance but with extremely high turnover, resulting in negative performance net of transaction costs. In contrast, slower-moving seasonal patterns, such as day-of-quarter, day-of-half-year and day-of-year, produce positive gross performance and, with the exception of day-of-quarter, positive net performance. An equal-weight ensemble of seasonality models produces an out-of-sample Sharpe ratio that is competitive with, and uncorrelated to, established global time series factor portfolios. Empirical results demonstrate that adding seasonality to a global factor portfolio increases the out-of-sample Sharpe ratio by approximately $30\%$ over the full evaluation period.

q-fin.MF↗

A Comparison for Non-Specialists of Workflow Steps and Similarity of Factor Rankings for Several Global Sensitivity Analysis Methods

Global sensitivity analysis (GSA) is a recommended step in the use of computer simulation models. GSA quantifies the relative importance of model inputs on outputs (Factor Ranking), identifies inputs that could be fixed, thus simplifying model calibration (Factor Fixing), and pinpointing areas for future data collection (Factor Prioritization). Given the wide variety of GSA methods, choosing between methods can be challenging for non-GSA experts. Issues include workflow steps and complexity, interpretation of GSA outputs, and the degree of similarity between methods in Factor Ranking. We conducted a study of both widely and less commonly used GSA methods applied to three simulators of differing complexity. All methods share common issues around implementation with specification of parameter ranges particularly critical. Similarities in Factor Rankings were generally high based on Kendall's W. Sobol' first order and total sensitivity indices were easy to interpret and informative with regression trees providing additional insight into interactions.

stat.ME↗

Causes of evolutionary divergence in prostate cancer

Cancer progression involves the sequential accumulation of genetic alterations that cumulatively shape the tumour phenotype. In prostate cancer, tumours can follow divergent evolutionary trajectories that lead to distinct subtypes, but the causes of this divergence remain unclear. While causal inference could elucidate the factors involved, conventional methods are unsuitable due to the possibility of unobserved confounders and ambiguity in the direction of causality. Here, we propose a method that circumvents these issues and apply it to genomic data from 829 prostate cancer patients. We identify several genetic alterations that drive divergence as well as others that prevent this transition, locking tumours into one trajectory. Further analysis reveals that these genetic alterations may cause each other, implying a positive-feedback loop that accelerates divergence. Our findings provide insights into how cancer subtypes emerge and offer a foundation for genomic surveillance strategies aimed at monitoring the progression of prostate cancer.

q-bio.GN↗

Effects of Distributed Roughness on Crossflow Instability through Generalized Resonance Mechanisms

Experiments have shown that micron-sized distributed surface roughness can significantly promote transition in a three-dimensional boundary layer dominated by crossflow insta- bility. This sensitive effect has not yet been fully explained physically and mathematically. Unlike past researches focusing on the receptivity of the boundary layer to surface roughness, or on the local stability of the modified mean flow, this paper seeks possible inherent mechanisms by investigating the effects of distributed surface roughness on crossflow instability through resonant interactions with eigenmodes. A key observation is that the perturbation induced by roughness with specific wavenumbers can interact with two eigenmodes (travelling and stationary vortices) through triadic resonance, or interact with one eigenmode (stationary vortices) through Bragg scattering. Unlike the usual triadic resonance of neutral, or nearly neutral, eigenmodes, the present triadic resonance can take place among modes with O(1) growth rates, provided that these are equal; unlike the usual Bragg scattering involving neutral waves, crossflow stationary vortices can also be unstable. For these amplifying waves, the generalized triadic resonance and Bragg scattering are put forward, and the resulting corrections to the growth rates are derived by a multiple-scale method. The analysis is extended to the case where up to four crossflow vortices interact with each other in the presence of suitable roughness components. The numerical results for Falkner-Skan-Cooke boundary layers show that roughness with a small height (a few percent of the local boundary-layer thickness) can change growth rates substantially (by a more-or-less O(1) amount)....

physics.flu-dyn↗

Extreme Events of Markov Chains

The extremal behaviour of a Markov chain is typically characterized by its tail chain. For asymptotically dependent Markov chains existing formulations fail to capture the full evolution of the extreme event when the chain moves out of the extreme tail region and for asymptotically independent chains recent results fail to cover well-known asymptotically independent processes such as Markov processes with a Gaussian copula between consecutive values. We use more sophisticated limiting mechanisms that cover a broader class of asymptotically independent processes than current methods, including an extension of the canonical Heffernan-Tawn normalization scheme, and reveal features which existing methods reduce to a degenerate form associated with non-extreme states.

math.PR↗