Equities

HARd to Beat: The Overlooked Impact of Rolling Windows in the Era of Machine Learning ArXiv ID: 2406.08041 “View on arXiv” Authors: Unknown Abstract We investigate the predictive abilities of the heterogeneous autoregressive (HAR) model compared to machine learning (ML) techniques across an unprecedented dataset of 1,455 stocks. Our analysis focuses on the role of fitting schemes, particularly the training window and re-estimation frequency, in determining the HAR model’s performance. Despite extensive hyperparameter tuning, ML models fail to surpass the linear benchmark set by HAR when utilizing a refined fitting approach for the latter. Moreover, the simplicity of HAR allows for an interpretable model with drastically lower computational costs. We assess performance using QLIKE, MSE, and realized utility metrics, finding that HAR consistently outperforms its ML counterparts when both rely solely on realized volatility and VIX as predictors. Our results underscore the importance of a correctly specified fitting scheme. They suggest that properly fitted HAR models provide superior forecasting accuracy, establishing robust guidelines for their practical application and use as a benchmark. This study not only reaffirms the efficacy of the HAR model but also provides a critical perspective on the practical limitations of ML approaches in realized volatility forecasting. ...

Modeling a Financial System with Memory via Fractional Calculus and Fractional Brownian Motion ArXiv ID: 2406.19408 “View on arXiv” Authors: Unknown Abstract Financial markets have long since been modeled using stochastic methods such as Brownian motion, and more recently, rough volatility models have been built using fractional Brownian motion. This fractional aspect brings memory into the system. In this project, we describe and analyze a financial model based on the fractional Langevin equation with colored noise generated by fractional Brownian motion. Physics-based methods of analysis are used to examine the phase behavior and dispersion relations of the system upon varying input parameters. A type of anomalous marginal glass phase is potentially seen in some regions, which motivates further exploration of this model and expanded use of phase behavior and dispersion relation methods to analyze financial models. ...

Interconnected Markets: Exploring the Dynamic Relationship Between BRICS Stock Markets and Cryptocurrency ArXiv ID: 2406.07641 “View on arXiv” Authors: Unknown Abstract This study aims to examine the intricate dynamics between BRICS traditional stock assets and the evolving landscape of cryptocurrencies. Using a time-varying parameter vector autoregression model (TVP-VAR), we have analyzed data from the BRICS stock market index, cryptocurrencies, and indicators from January 6, 2015, to June 29, 2023. The results show that three out of the five BRICS stock markets serve as primary sources of shocks that subsequently affect the financial network. The transcontinental (TCI) value derived from the dynamic conditional connectedness using the TVP-VAR model demonstrates a higher explanatory power than the static connectedness observed using the standard VAR model. The discoveries from this study offer valuable insights for corporations, investors, and regulators concerning systematic risk and investment strategies. ...

Application of Black-Litterman Bayesian in Statistical Arbitrage ArXiv ID: 2406.06706 “View on arXiv” Authors: Unknown Abstract \begin{“abstract”} In this paper, we integrated the statistical arbitrage strategy, pairs trading, into the Black-Litterman model and constructed efficient mean-variance portfolios. Typically, pairs trading underperforms under volatile or distressed market condition because the selected asset pairs fail to revert to equilibrium within the investment horizon. By enhancing this strategy with the Black-Litterman portfolio optimization, we achieved superior performance compared to the S&P 500 market index under both normal and extreme market conditions. Furthermore, this research presents an innovative idea of incorporating traditional pairs trading strategies into the portfolio optimization framework in a scalable and systematic manner. ...

Stock Movement Prediction with Multimodal Stable Fusion via Gated Cross-Attention Mechanism ArXiv ID: 2406.06594 “View on arXiv” Authors: Unknown Abstract The accurate prediction of stock movements is crucial for investment strategies. Stock prices are subject to the influence of various forms of information, including financial indicators, sentiment analysis, news documents, and relational structures. Predominant analytical approaches, however, tend to address only unimodal or bimodal sources, neglecting the complexity of multimodal data. Further complicating the landscape are the issues of data sparsity and semantic conflicts between these modalities, which are frequently overlooked by current models, leading to unstable performance and limiting practical applicability. To address these shortcomings, this study introduces a novel architecture, named Multimodal Stable Fusion with Gated Cross-Attention (MSGCA), designed to robustly integrate multimodal input for stock movement prediction. The MSGCA framework consists of three integral components: (1) a trimodal encoding module, responsible for processing indicator sequences, dynamic documents, and a relational graph, and standardizing their feature representations; (2) a cross-feature fusion module, where primary and consistent features guide the multimodal fusion of the three modalities via a pair of gated cross-attention networks; and (3) a prediction module, which refines the fused features through temporal and dimensional reduction to execute precise movement forecasting. Empirical evaluations demonstrate that the MSGCA framework exceeds current leading methods, achieving performance gains of 8.1%, 6.1%, 21.7% and 31.6% on four multimodal datasets, respectively, attributed to its enhanced multimodal fusion stability. ...

Portfolio Optimization with Robust Covariance and Conditional Value-at-Risk Constraints ArXiv ID: 2406.00610 “View on arXiv” Authors: Unknown Abstract The measure of portfolio risk is an important input of the Markowitz framework. In this study, we explored various methods to obtain a robust covariance estimators that are less susceptible to financial data noise. We evaluated the performance of large-cap portfolio using various forms of Ledoit Shrinkage Covariance and Robust Gerber Covariance matrix during the period of 2012 to 2022. Out-of-sample performance indicates that robust covariance estimators can outperform the market capitalization-weighted benchmark portfolio, particularly during bull markets. The Gerber covariance with Mean-Absolute-Deviation (MAD) emerged as the top performer. However, robust estimators do not manage tail risk well under extreme market conditions, for example, Covid-19 period. When we aim to control for tail risk, we should add constraint on Conditional Value-at-Risk (CVaR) to make more conservative decision on risk exposure. Additionally, we incorporated unsupervised clustering algorithm K-means to the optimization algorithm (i.e. Nested Clustering Optimization, NCO). It not only helps mitigate numerical instability of the optimization algorithm, but also contributes to lower drawdown as well. ...

Optimizing Broker Performance Evaluation through Intraday Modeling of Execution Cost ArXiv ID: 2405.18936 “View on arXiv” Authors: Unknown Abstract Minimizing execution costs for large orders is a fundamental challenge in finance. Firms often depend on brokers to manage their trades due to limited internal resources for optimizing trading strategies. This paper presents a methodology for evaluating the effectiveness of broker execution algorithms using trading data. We focus on two primary cost components: a linear cost that quantifies short-term execution quality and a quadratic cost associated with the price impact of trades. Using a model with transient price impact, we derive analytical formulas for estimating these costs. Furthermore, we enhance estimation accuracy by introducing novel methods such as weighting price changes based on their expected impact content. Our results demonstrate substantial improvements in estimating both linear and impact costs, providing a robust and efficient framework for selecting the most cost-effective brokers. ...

Exploring Sectoral Profitability in the Indian Stock Market Using Deep Learning ArXiv ID: 2407.01572 “View on arXiv” Authors: Unknown Abstract This paper explores using a deep learning Long Short-Term Memory (LSTM) model for accurate stock price prediction and its implications for portfolio design. Despite the efficient market hypothesis suggesting that predicting stock prices is impossible, recent research has shown the potential of advanced algorithms and predictive models. The study builds upon existing literature on stock price prediction methods, emphasizing the shift toward machine learning and deep learning approaches. Using historical stock prices of 180 stocks across 18 sectors listed on the NSE, India, the LSTM model predicts future prices. These predictions guide buy/sell decisions for each stock and analyze sector profitability. The study’s main contributions are threefold: introducing an optimized LSTM model for robust portfolio design, utilizing LSTM predictions for buy/sell transactions, and insights into sector profitability and volatility. Results demonstrate the efficacy of the LSTM model in accurately predicting stock prices and informing investment decisions. By comparing sector profitability and prediction accuracy, the work provides valuable insights into the dynamics of the current financial markets in India. ...

DSPO: An End-to-End Framework for Direct Sorted Portfolio Construction ArXiv ID: 2405.15833 “View on arXiv” Authors: Unknown Abstract In quantitative investment, constructing characteristic-sorted portfolios is a crucial strategy for asset allocation. Traditional methods transform raw stock data of varying frequencies into predictive characteristic factors for asset sorting, often requiring extensive manual design and misalignment between prediction and optimization goals. To address these challenges, we introduce Direct Sorted Portfolio Optimization (DSPO), an innovative end-to-end framework that efficiently processes raw stock data to construct sorted portfolios directly. DSPO’s neural network architecture seamlessly transitions stock data from input to output while effectively modeling the intra-dependency of time-steps and inter-dependency among all tradable stocks. Additionally, we incorporate a novel Monotonical Logistic Regression loss, which directly maximizes the likelihood of constructing optimal sorted portfolios. To the best of our knowledge, DSPO is the first method capable of handling market cross-sections with thousands of tradable stocks fully end-to-end from raw multi-frequency data. Empirical results demonstrate DSPO’s effectiveness, yielding a RankIC of 10.12% and an accumulated return of 121.94% on the New York Stock Exchange in 2023-2024, and a RankIC of 9.11% with a return of 108.74% in other markets during 2021-2022. ...

Data-generating process and time-series asset pricing ArXiv ID: 2405.10920 “View on arXiv” Authors: Unknown Abstract We study the data-generating processes for factors expressed in return differences, which the literature on time-series asset pricing seems to have overlooked. For the factors’ data-generating processes or long-short zero-cost portfolios, a meaningful definition of returns is impossible; further, the compounded market factor (MF) significantly underestimates the return difference between the market and the risk-free rate compounded separately. Surprisingly, if MF were treated coercively as periodic-rebalancing long-short (i.e., the same as size and value), Fama-French three-factor (FF3) would be economically unattractive for lacking compounding and irrelevant for suffering from the small “size of an effect.” Otherwise, FF3 might be misspecified if MF were buy-and-hold long-short. Finally, we show that OLS with net returns for single-index models leads to inflated alphas, exaggerated t-values, and overestimated Sharpe ratios (SR); worse, net returns may lead to pathological alphas and SRs. We propose defining factors (and SRs) with non-difference compound returns. ...