Short Portfolio Research Articles

An Empirical Study on the Impact of Investor Gambling Factors on Stock Cross-Sectional Returns

This study explores how investors' gambling preferences affect the cross-sectional returns of stocks in China's A-share market through behavioral finance factors. The portfolio ranking method and Fama-Macbeth regression analysis show how gambling behavior affects stock returns. On this basis, a multi-factor model (CHG model) incorporating gambling factor innovation was constructed to capture market dynamics more accurately. Research shows that going long on a low gaming index stock portfolio and short on a high gaming index stock portfolio can achieve significant average monthly returns. After adjustment by the Fama-French three-factor model, the returns of this long and short portfolio are significant at the 1% level. Significantly. In the CHG model, the coefficient of the gambling index Gamble is -0.0241, and the t-statistic is -6.1858, which is significant at the 1% level, further confirming the negative impact of gambling preferences on stock returns. In addition, the CHG model's ability to explain anomalies is also better than the CH-4 model. This study provides a new perspective on understanding behavioral finance factors in the Chinese stock market, and the results highlight the importance of considering behavioral finance factors in asset pricing and investment decisions in a highly irrational market environment. Future research can further explore the performance of behavioral financial factors in different market environments, as well as the differences and similarities of behavioral factors in different markets.

Factor timing in the Chinese stock market

AbstractI conduct an exploratory study about the feasibility of factor timing in the Chinese stock market, covering 24 representative and well‐identified risk factors in 10 categories from the literature. The long–short portfolio of short‐term reversal exhibits strong out‐of‐sample predictability, which is robust across various models and all types of predictors. This predictability is significant both statistically and economically, with a simple investment strategy obtaining its return three times higher than the buy‐and‐hold return in the sample period and a significant annualized 20.4% CH‐3 alpha. Portfolio historical volatility and market volatility measurement predictors play crucial roles in the reversal factor premium's robust predictability. However, such results are not evident in predicting all other factors' long–short portfolios as well as all factors' long‐wing and short‐wing portfolios, and this failure cannot be attributed to their exposure to unpredictable market returns.

An adaptive evolutionary strategy for long–short portfolio construction

An adaptive evolutionary strategy for long–short portfolio construction

Factor returns and FOMC announcements: The role of sentiment

We examine the dynamics of long-short factor returns on FOMC announcement days and the role of sentiment. We find that factor returns are negative on FOMC announcement days. Moreover, on these days returns are significantly lower following low sentiment periods. Hence, investor sentiment is a key driver of factor returns on FOMC days and this effect emanates mainly from the short portfolio leg of each factor.

Informed trading by hedge funds

AbstractUsing daily equity transactions, we create a hedge fund informed trading measure (ITM) that separates concentrated information‐related trades from liquidity‐driven basket trades. We find that stocks with higher ITM are associated with higher future stock performance. The long–short portfolio delivers 4% annual alpha after controlling for size, value, momentum, and illiquidity factors. We attribute informed trading to hedge funds' ability to identify and correct stock underpricing. The results are robust to several ways of constructing and sorting the measure, and we do not find a return reversal in four quarters, indicating that the measure is information related.

Outperforming the market: a comparison of Star and NonStar analysts’ investment strategies and recommendations

AbstractWe employ StarMine to investigate the impact of analyst recommendations on stock performance. We test whether star-ranked analysts generate abnormal returns and outperform non-stars in short and long portfolios. Utilizing buy-and-hold calendar-time portfolio methodology, we calculate portfolio alphas using various asset pricing models, including CPM, the Fama and French 3-factor model, and the Carhart 4-factor model. Results indicate that all analyst groups can generate abnormal returns exceeding the market average. Star-ranked analysts outperform non-stars in short portfolios by 0.5523% in monthly alpha, though no significant difference exists in long portfolio alphas. We also conduct regressor endogeneity tests and explore investor sentiment mechanisms by utilizing the GARCH model and frequency-domain causality analysis, with NASDAQ as a proxy for investor sentiment. These tests reveal that the momentum factor is exogenous, and investor sentiments have a statistically significant positive effect on stock return volatility, with changes occurring between 5 and 10 days. This research underscores the value of analyst insights for investors, validates StarMine’s ranking effectiveness, and suggests market participants can benefit from incorporating analyst recommendations into their investment decisions. Our study makes a significant contribution to the existing literature by introducing a novel approach to understanding investor sentiment mechanisms through a causality model.

Predictability of commodity futures returns with machine learning models

AbstractWe use prevailing machine learning models to investigate the predictability of futures returns in 22 commodities with commodity‐specific and macroeconomic factors as predictors. Out‐of‐sample prediction errors for the majority of futures contracts are lowered compared with those obtained by the baseline models of AR(1) and forecast combinations. Using Shapley values to explain feature importance, we identify dominant predictors for each commodity. A long–short portfolio strategy based on monthly light gradient‐boosting machine predictions outperforms the benchmark linear models in terms of annual return, Sharpe ratio, and max drawdown.

인공신경망(ANN)을 이용한 옵션가격결정 및 딥헤징 (DH)에 대한 SNN 및 RNN 결과의 비교·분석

The purpose of this study is to present alternative methods to overcome market limitations such as the assumption of fixing the dynamics of underlying assets and market friction for the traditional Black-Scholes option pricing model. As for the research method of this paper, Adam is used as the gradient descent optimizer for the hedging model of the call option short portfolio using artificial neural network models SNN (simple neural network) and RNN (recurrent neural network) as analysis models, and deep neural network (deep neural network) is used as the hedging model. neural network) methodology was applied.
 The research results of this paper are as follows. Regarding the ATM call option price, the BS model showed 10.245, the risk neutral model showed 10.268, the SNN-DH model showed 11.834, and the RNN-DH model showed 11.882. Therefore, it appears that there is a slight difference in the call option price according to each analysis model. At this time, the DH analysis data sample is 100,000 (1×) training samples generated by Monte Carlo Simulation, and the number of testing samples is the training sample. 20% (20,000 pieces) was used, and the option payoff function is lambda(), which is -max(, 0). In addition, the option price and P&L (P&L) of SNN-DH and RNN-DH appear linearly on the basis, and the longer the remaining maturity, the closer the delta value of SNN-DH and BS to the distribution of the S-curve, and the closer the expiration date, the closer the two models are. Delta is densely distributed between 0 and 1. As for the total loss of delta hedging of the short call option position, SNN-DH was -0.0027 and RNN-DH was -0.0061, indicating that the total hedged profit and loss was close to 0.
 The implications of this study are to present DL techniques based on AI methods as an alternative way to overcome limitations such as fixing underlying asset dynamics and market friction of the traditional Black-Scholes model. In addition, it is an independent analysis model that considers valid robust tools by applying deep neural network algorithms for portfolio hedging problems. One limitation of the research is that it analyzed the model under the assumption of zero transaction costs, so future studies should consider this aspect.

Exploring Low-Risk Anomalies: A Dynamic CAPM Utilizing a Machine Learning Approach

Low-risk pricing anomalies, characterized by lower returns in higher-risk stocks, are prevalent in equity markets and challenge traditional asset pricing theory. Previous studies primarily relied on linear regression methods, which analyze a limited number of factors and overlook the advantages of machine learning in handling high-dimensional data. This study aims to address these anomalies in the Chinese market by employing machine learning techniques to measure systematic risk. A large dataset consisting of 770 variables, encompassing macroeconomic, micro-firm, and cross-effect factors, was constructed to develop a machine learning-based dynamic capital asset pricing model. Additionally, we investigated the differences in factors influencing time-varying beta between state-owned enterprises (SOEs) and non-SOEs, providing economic explanations for the black-box issues. Our findings demonstrated the effectiveness of random forest and neural networks, with the four-layer neural network performing best and leading to a substantial rise in the excess return of the long–short portfolio, up to 0.36%. Notably, liquidity indicators emerged as the primary drivers influencing beta, followed by momentum. Moreover, our analysis revealed a shift in variable importance during the transition from SOEs to non-SOEs, as liquidity and momentum gradually replaced fundamentals and valuation as key determinants. This research contributes to both theoretical and practical domains by bridging the research gap in incorporating machine learning methods into asset pricing research.

Big data analytics and investment

Big data has found extensive applications in various industries, including finance. It is an essential tool for investors to make high-stakes investment decisions. Using China’s A-shares Market, this paper employs 76 firm characteristics to conduct descriptive analytics (factor model) and predictive analytics (long–short portfolio) through an Instrumented Principal Component Analysis (IPCA) model. According to our results, the IPCA model outperforms in both description (tangency portfolio Sharpe ratio of 2.91) and forecasting (long–short portfolio Sharpe ratio of 2.38). Moreover, our paper compares the performance of different sets of characteristics in big data analytics and concludes that sentiment is dominant, while fundamental analysis is also important. Our results can provide policymakers with valuable insights into the common trends of the stock market and assist investors in making effective investment decisions.

Automated stock picking using random forests

We derive a stock ranking by applying a technical features-based random forest model on an international dataset of liquid stocks. Rather than predicted return, our ranking is based on outperformance probability. By applying a decile split, we find that long–short portfolios achieve Sharpe ratios of up to 1.95 and a highly significant yearly six-factor alpha of up to 21.79%. Moreover, we show that outperformance probabilities serve as a superior measure of future returns in the context of portfolio optimization. Mean–variance portfolios using this measure are less volatile and more profitable than equally- or value-weighted portfolios. Our findings are robust to firm size, regional restrictions, and non-crisis periods and cannot be explained by limits to arbitrage.

Recommendation Algorithm of Industry Stock Trading Model with TODIM

In stock trading, a common phenomenon is that the trends of stocks in the same industry are very similar. In contrast, the movements of stocks in different industries are often different. Therefore, applying the same model to all stock trading is inappropriate without distinguishing the industries in which the stocks belong. However, recommending an optimal industry stock trading model is very challenging based on performance evaluation indicators. First, the indicators of the trading model are diverse. Second, the ranking of multiple indicators is often inconsistent. In the paper, we model the problem to be solved as a multi-criteria decision-making process. Therefore, we first divide stock dataset into nine industries according to their main business. Then, we apply several machine learning algorithms as candidate models to generate trading signals. Second, we conduct daily trading backtesting based on the trading signals to obtain multiple performance evaluation indicators. Third, we propose an optimal recommendation algorithm for the industry stock trading model with TODIM. The experimental results in the US stock market and China’s A-share market show that the proposed algorithm can get a better trading model out-of-sample industry stock. Moreover, we effectively evaluate the generalization ability of the algorithm based on the proposed metrics. Finally, the proposed long–short portfolios based on the algorithm have achieved returns exceeding the benchmark on most out-of-sample datasets.

Practical Applications of Questioning the Wisdom of Crowds to Design Portfolio Diversification Strategies

In <b><i>Questioning the Wisdom of Crowds to Design Portfolio Diversification Strategies</i></b>, from the Winter 2023 issue of <b><i>The Journal of Alternative Investments</i></b>, <b>Vadim Zlotnikov</b>, <b>Igor Halperin</b>, <b>Lisa Huang</b>, <b>Cathy Pena</b> (all of <b>Fidelity</b>), and <b>Mikhail Stukalo</b> (of <b>Gainy</b>, a fintech startup), develop a diversification strategy based on measures of crowdedness. They design a portfolio with the positive characteristics of uncrowded stocks that also harvests the skill of active managers. They create a measure of crowdedness using holdings data on active managers. Stocks can become crowded based on conviction, reflecting manager skill. But they also can become crowded due to a consensus view, which leads to problematic characteristics; crowded stocks can exhibit negatively skewed returns with high forward-looking volatility. The authors also construct a long–short portfolio based on two different measures of crowdedness. The portfolio is beta-neutral and can be used to improve the Sharpe ratio of a 60/40 portfolio by dampening volatility. Maintaining smart money investors’ insights, the equity component does not sacrifice returns, as other defensive strategies do.

Emerging Markets Currency Factors and US High-Frequency Macroeconomic Shocks

Relying on the structural vector autoregression developed by Cieslak and Pang, the authors identify four shocks to the US economy based on the US Treasury yield curve and the stock market: two fundamental news shocks (growth and money) and two risk-premium shocks (common and hedging). They find that these shocks explain over 40% of the time-series variation of emerging markets currency (EMFX) returns. Additionally, EMFX returns increase significantly with positive growth shocks and decrease with monetary tightening and risk-premium shocks. The authors show that growth and common shocks are priced in the cross section of EMFX, with a positive and negative risk premiums, respectively. They then build long–short currency portfolios based on several academically researched style factors and test their performance and relative exposure to the macroeconomic shocks affecting the US economy. They find that only carry and macro momentum long–short portfolios generate positive and significant alphas and excess returns over their sample. However, all single-factor portfolios have sizable exposure to the four shocks. The authors show that a simple multifactor approach to investing in EMFXs eliminates the exposure of excess returns to all macroeconomic shocks.

Consumer Spending and the Cross-Section of Stock Returns

Using a unique dataset of individual transactions-level data for a universe of US consumer facing stocks, we examine the information content of consumer credit and debit card spending in explaining future stock returns. Our analysis shows that consumer spending data positively predict various measures of a company’s future earnings surprises up to three quarters in the future. This predictive power remains strong in both large- and small-cap universes of consumer discretionary firms in our sample and is robust to the type of transactions data considered (credit card, debit card, or both), although the relationship is stronger in the small-cap universe where informational asymmetries are more pronounced. Based on this empirical observation, we build a simple long–short strategy that takes long–short positions in the top/bottom tercile of stocks ranked on our real-time sales signal. The strategy generates statistically and economically significant returns of 16% per annum net of transaction costs and after controlling for the common sources of systematic factor returns. A simple optimization exercise to form (tangency) mean–variance-efficient portfolios of factors leads to an optimal factor allocation that assigns almost 50% weight to our long–short portfolio. Our results suggest that consumer transaction level data can serve as a more accurate and persistent signal of a firm’s growth potential and future returns.

The strategic allocation to style-integrated portfolios of commodity futures

Our study lies at the intersection of the literature on the diversification benefits of commodity futures and the literature on style integration. It augments the traditional asset mix of investors with a long–short portfolio that integrates the styles that matter to the pricing of commodity futures. Treating the style-integrated portfolio of commodities as part of the strategic mix of investors is found to enhance out-of-sample performance and reduce crash risk compared to the alternatives considered thus far. The conclusion holds across traditional asset mix, portfolio allocation methods, integration strategies, and sub-periods. The diversification benefits of style integration also persist, albeit lower, in a long-only setting.

A large-dimensional test for cross-sectional anomalies:Efficient sorting revisited

Many researchers seek factors that predict the cross-section of stock returns. In finance, the key is to replicate anomalies by long–short portfolios based on their firm characteristics, with microcap biases alleviated via New York Stock Exchange (NYSE) breakpoints and value-weighted returns. In econometrics, the key is to include a covariance matrix estimator of stock returns for (mimicking) the portfolio construction. This paper marries these two strands of literature in order to test the zoo of cross-sectional anomalies by injecting size controls, basically NYSE breakpoints and value-weighted returns, into efficient sorting. We propose to use a covariance matrix estimator for ultra-high dimensions (up to 5,000) taking into account large, small and microcap stocks. We demonstrate that using a nonlinear shrinkage estimator of the covariance matrix substantially enhances the power of tests for cross-sectional anomalies: On average, t-statistics more than double. Furthermore, the proposed revisited efficient sorting method computes even highly significant factor portfolios net of transaction costs.

Mean–Variance Portfolio Efficiency under Leverage Aversion and Trading Impact

This paper addresses the optimal rebalancing problem of a long–short portfolio with high net asset value under trading impact losses. The fund manager may employ leveraging as a tool to increase portfolio returns. However, to mitigate potential leverage risks, frequent rebalancing may become necessary, which leads to significant slippage losses that dampen portfolio performance ex post. We consider the problem in an integrated framework by incorporating trading impact and leverage restrictions ex ante within a mean–variance framework, where leverage control is imposed using a chance constraint. The resulting mean–variance–leverage optimization model (MVL) is non-convex, and we develop an efficient scheme to obtain the optimal portfolio. We investigate how portfolio leverage modifies the MV efficient frontier in the presence of trading impact, and highlight the significant outperformance of the proposed model relative to the standard mean–variance model. Increased target means require less restrictions on leverage, which result in higher rates of slippage losses. Our analysis supports the notion that leverage restrictions contribute to choosing high beta assets, even in the presence of trading impact.

Machine learning portfolios with equal risk contributions: Evidence from the Brazilian market

We investigate the use of machine learning (ML) to forecast stock returns in the Brazilian market using a rich proprietary dataset. While ML portfolios can easily outperform the local market, the performance of long-short strategies using ML is hampered by the high volatility of the short portfolios. We show that an Equal Risk Contribution (ERC) approach significantly improves risk-adjusted returns. We further develop an ERC approach that combines multiple long-short strategies obtained with ML models, equalizing risk contributions across ML models, which outperforms, on a risk-adjusted basis, all individual ML long-short strategies, as well as alternative combinations of ML strategies.

Tuning Trend-Following Strategies with Macro ESG Data

This article seeks to tilt traditional macro trend-following strategies toward countries with high Environmental, Social, and Governance (ESG) scores. The integration incorporates both ESG levels and changes (improvements or deteriorations in sustainability). Notably, the authors find that the international ESG exposure of the macro portfolios can be substantially increased without any cost in performance for both long-only and long–short portfolios. In some cases, transaction cost–adjusted Sharpe ratios can even benefit from a minor shift toward more ESG exposure.