#869 – Market Timing Corporate Bonds with Machine learning \u2013 Random Forests<\/a><\/strong><\/p>\n\n\n\n Period of rebalancing:<\/strong> Monthly Source paper:<\/strong><\/p>\n\n\n\n He, X. and Feng, G. and Wang, J. and Wu, CH.: Predicting Individual Corporate Bond Returns<\/strong> #870 – Timing Carry Trade with Central Banks\u2019 Announcements<\/a><\/strong><\/p>\n\n\n\n Period of rebalancing:<\/strong> Daily Source paper:<\/strong><\/p>\n\n\n\n B\u00f6ck (Boeck), Maximilian and Steshkova, Alina and Zoerner (Z\u00f6rner), Thomas O.: The Impact of Currency Carry Trade Activity on the Transmission of Monetary Policy<\/strong> #871 – Traditional Carry in Cryptocurrencies<\/a><\/strong><\/p>\n\n\n\n Period of rebalancing:<\/strong> Weekly Source paper:<\/strong><\/p>\n\n\n\n Fan, Zhenzhen and Jiao, Feng and Lu, Lei and Tong, Xin: Risk-Return Relation of Cryptocurrency Carry Trade<\/strong> #872 – Machine Learning \u2013 Random Forrests Predicts Cross Section of Corporate Bonds<\/a><\/strong><\/p>\n\n\n\n Period of rebalancing:<\/strong> Monthly Source paper:<\/strong><\/p>\n\n\n\n He, X. and Feng, G. and Wang, J. and Wu, CH.: Predicting Individual Corporate Bond Returns<\/strong> #873 – Timing the Factor Zoo via Deep Learning: Evidence from China<\/a><\/strong><\/p>\n\n\n\n Period of rebalancing:<\/strong> Daily Source paper:<\/strong><\/p>\n\n\n\n Ma, T. and Liao, C. and Jiang, F.: Timing the Factor Zoo via Deep Learning: Evidence from China<\/strong> #874 – Commodity Pairs Trading in China using Machine Learning<\/a><\/strong><\/p>\n\n\n\n Period of rebalancing:<\/strong> Daily Source paper:<\/strong><\/p>\n\n\n\n Huang, K., Sun, J., Zhang, Z. and Li, Q.: Pair trading of Commodity Futures in China through the lens of intraday data A Machine Learning Framework and Empirical Analysis<\/strong> #464 – Brand Value Asset Pricing Factor<\/a><\/strong><\/p>\n\n\n\n Fischer, Stefan and Weiger, Welf and Hammerschmidt, Maik: Same Same but Different? The Predictive Power of Association Types in Brand Buzz for Investor Returns <\/strong>https:\/\/ssrn.com\/abstract=4336934<\/a> #460 – ESG Level Factor Investing Strategy<\/a><\/strong><\/p>\n\n\n\n Capot\u0103, Laura-Dona and Giuzio, Margherita and Kapadia, Sujit and Salakhova, Dilyara: Are Ethical and Green Investment Funds More Resilient?<\/strong> Tham, Eric: Greenwashing Premium<\/strong> #839 – Factor Allocation with Reinforcement Learning<\/a><\/strong><\/p>\n\n\n\n Lavko, Matus and Klein, Tony and Walther, Thomas: Reinforcement Learning and Portfolio Allocation: Challenging Traditional Allocation Methods<\/strong> #536 – Machine Learning Stock Picking<\/a><\/strong><\/p>\n\n\n\n Lin, Weidong and Taamouti, Abderrahim: Machine Learning Based Portfolio Selection Under Systemic Risk #827 – A Machine Learning Approach to Stock Returns Prediction in China<\/a><\/strong><\/p>\n\n\n\n Wang, Yonghao and Peng, Qinke and Han, Tian and Li, Haozhou and Shen, Yiqing: Deep Reinforcement Learning Based End-to-End Stock Trading Strategy<\/strong> #810 – Integrating ESG into Fixed Income Portfolios<\/a><\/strong><\/p>\n\n\n\n Lei, Ivy Qiongwen: Predicting Corporate Governance Effect on Corporate Bond Yields Using Neural Networks<\/strong> Comparison of Commodity Momentum Strategy in the U.S. and Chinese Markets<\/strong><\/a><\/p>\n\n\n\n The commodity momentum strategy is a crucial driving force behind Commodity Trading Advisor (CTA) strategies, as it capitalizes on the persistence of price trends in various commodity markets. By identifying and exploiting these trends, CTAs can achieve robust returns and diversification benefits. In their new paper, John Hua FAN and Xiao QIAO (February 2023) present their perspective and understanding of cross-country and cross-sector influences on the behavior of commodity momentum beyond established commodity fundamentals focusing on U.S. and China markets.<\/p>\n\n\n\n <\/a>How to Rebalance Smart Beta Strategies Smarter<\/strong><\/a><\/p>\n\n\n\n The topic of Smart-Beta is widely recognized, and we cover, monitor, and inform about its developments. The analyzed piece is about the importance of the correct rebalancing strategy and is kindly provided by Research Affiliates. According to a recent research article, investors should re-consider rebalancing with turnover constraint only those stocks that have the strongest signal. Prioritizing trades in stocks that are the farthest removed from the portfolio selection threshold is likely to minimize the expected need for additional trading.<\/p>\n\n\n\n Anomaly Discovery and Arbitrage Trading<\/strong><\/a><\/p>\n\n\n\n Today, we will look closer into the hood of life expectancy of investment strategies and try to answer the critical question on which many, in some sense, if not all, trading strategies are built: what happens with anomalies after their discovery? The paper\u2019s authors, with the sweet, simple name Anomaly Discovery and Arbitrage Trading, analyze a stylized model of anomaly discovery, which has implications for both asset prices and arbitrageurs\u2019 trading. Their original research produced an arbitrageur-based asset pricing model that shows that discovering an anomaly reduces the correlation between the returns of its long- and short-leg portfolios: HFs (professional arbitrageurs) use to increase (unwind) such trades when their wealth increases (decreases), further supporting the view that the discovery effects work through arbitrage trading. This effect is more substantial when arbitrageurs\u2019 wealth is more volatile.<\/a><\/p>\n\n\n\n #192 – Complexity Effect in Stocks<\/a> Six new strategies have been added. Seven new related research papers have been included into existing strategy reviews and three short free blog posts<\/strong><\/a> has been published during last few weeks. Plus, four trading strategies have been backtested in QuantConnect<\/strong><\/a> in the previous two weeks.<\/p>","protected":false},"author":22768,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-26824","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/posts\/26824","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/users\/22768"}],"replies":[{"embeddable":true,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/comments?post=26824"}],"version-history":[{"count":0,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/posts\/26824\/revisions"}],"wp:attachment":[{"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/media?parent=26824"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/categories?post=26824"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/tags?post=26824"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
Markets traded: <\/strong>bonds
Instruments used for trading:<\/strong> ETFs, funds
Complexity:<\/strong> Very complex strategy
Backtest period:<\/strong> 1973-2020
Indicative performance:<\/strong> 19.7%
Estimated volatility:<\/strong> 11.45%<\/p>\n\n\n\n
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=4374213<\/a>
Abstracto:
This paper documents substantial evidence of return predictability and investment gains for individual corporate bonds via machine learning. The forecast-implied long-short and market-timing strategies deliver significant risk-adjusted returns over transaction costs. Random Forest has the best performance as the ensemble of regression trees helps reduce overfitting. Using a long-span sample from 1976 to 2020, we can evaluate return predictability over business cycles and find aggregate predictors (e.g., corporate bond market return and TERM factor) that show higher forecasting power than bond characteristics. Finally, we find return predictability differs between bonds issued by private and public firms, with higher investment gains in private bonds.<\/p>\n\n\n\n
Markets traded: <\/strong>currencies
Instruments used for trading:<\/strong> CFDs, forwards, futures, swaps
Complexity:<\/strong> Moderately complex strategy
Backtest period:<\/strong> 2004-2019
Indicative performance:<\/strong> 4.47%
Estimated volatility:<\/strong> 4.88%<\/p>\n\n\n\n
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=4371901<\/a>
Abstracto:
In this paper, we examine how carry trade activity affects the transmission of monetary policy in currency markets. We analyze a large set of currencies against the U.S. dollar. The U.S. dollar appreciates in response to a conventional monetary policy shock but depreciates to an information shock. Currencies typically involved in the carry trade tend to respond stronger to both shocks, while safe-haven currencies exhibit different adjustment dynamics. To infer these effects from the data, a threshold vector autoregressive model is fitted to discriminate between different regimes of carry trade activity. Finally, a currency trading strategy created on the day of central bank announcements, which takes into consideration the joint co-movement of interest rates and stock prices, outperforms strategies based on carry trade or on the dollar risk factor in terms of the Sharpe ratio and downside risk.<\/p>\n\n\n\n
Markets traded: <\/strong>cryptos
Instruments used for trading:<\/strong> cryptos
Complexity:<\/strong> Simple strategy
Backtest period:<\/strong> 2016-2021
Indicative performance:<\/strong> 46.71%
Estimated volatility:<\/strong> 60.68%<\/p>\n\n\n\n
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=4397327<\/a>
Abstracto:
This paper comprehensively examines the risk-return relation of cryptocurrency carry trade using realistic borrowing and lending interest rates. We find significant violations of the uncovered interest rate parity in the cryptocurrency market. The cross-sectional carry trade strategy yields an annualized return of 46.71% and a Sharpe ratio of 0.77. Unlike fiat-currency carry trade which is vulnerable to crash risk, the cryptocurrency carry trade is resistant to the cryptocurrency market crashes in 2018 and 2021. We show that the crypto-carry trade returns cannot be explained by established risk factors from fiat currencies or cryptocurrencies. We find that geopolitical risk explains a substantial amount of the carry returns. Interestingly, the risk-adjusted crypto-carry return alpha is significantly negative, implying a negative cryptocurrency carry risk premium.<\/p>\n\n\n\n
Markets traded: <\/strong>bonds
Instruments used for trading:<\/strong> bonds
Complexity:<\/strong> Very complex strategy
Backtest period:<\/strong> 1973-2020
Indicative performance:<\/strong> 24.7%
Estimated volatility:<\/strong> 7.24%<\/p>\n\n\n\n
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=4374213<\/a>
Abstracto:
This paper documents substantial evidence of return predictability and investment gains for individual corporate bonds via machine learning. The forecast-implied long-short and market-timing strategies deliver significant risk-adjusted returns over transaction costs. Random Forest has the best performance as the ensemble of regression trees helps reduce overfitting. Using a long-span sample from 1976 to 2020, we can evaluate return predictability over business cycles and find aggregate predictors (e.g., corporate bond market return and TERM factor) that show higher forecasting power than bond characteristics. Finally, we find return predictability differs between bonds issued by private and public firms, with higher investment gains in private bonds.<\/p>\n\n\n\n
Markets traded: <\/strong>equities
Instruments used for trading:<\/strong> stocks
Complexity:<\/strong> Very complex strategy
Backtest period:<\/strong> 2004-2020
Indicative performance:<\/strong> 59%
Estimated volatility:<\/strong> 42%<\/p>\n\n\n\n
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=4247722<\/a>
Abstracto:
This paper proposes a factor timing strategy with information from 146 characteristic-based factors and a deep learning approach to capture nonlinear predictability. The deep learning-based factor timing strategy generates the highest economic value compared with the unconditional and alternative linear machine learning-based portfolios and remains robust after controlling for traditional factor models and transaction costs. With the unique market structure of the Chinese stock market, we find that mispricing-based theory helps explain the factor timing via deep learning.<\/p>\n\n\n\n
Markets traded: <\/strong>commodities
Instruments used for trading:<\/strong> CFDs, futures
Complexity:<\/strong> Very complex strategy
Backtest period:<\/strong> 2019-2021
Indicative performance:<\/strong> 13.84%
Estimated volatility:<\/strong> 7.64%<\/p>\n\n\n\n
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=4117105<\/a>
Abstracto:
In statistical arbitrage, paired trading, as a market-neutral strategy, is widely used because of its simple method and easy implementation. This paper constructs a machine learning framework for commodity futures matching trading from four aspects: systematic feature extraction, unsupervised learning, pairing filtering rules and algorithmatic trading based on spread prediction. This paper selects 47 highly liquid commodities in China\u2019s commodity futures market and establishes trading strategies using a continuous commodity index constructed from 5-minute high-frequency data within the 2019-2021 intraday. The results show that the machine learning framework selects potential matching combinations that cannot be identified by traditional commodity classification from all possible matching combinations. After taking transaction costs into account, the results of the out-of-sample experiment confirm that the machine learning framework\u2019s pair selection program can produce a more profitable and robust commodity arbitrage portfolio, with 1.8 Sharp and 13.8% annualized return, but the in-sample traditional commodity classification arbitrage is more profitable. When deep learning prediction model is introduced to compare with non-prediction and traditional linear prediction model, the traditional threshold model with no prediction of spread is superior to the prediction model, and LSTM prediction is superior to ARIMA prediction. The conclusion of this paper can provide important enlightenment for designing better statistical arbitrage strategies of commodity futures and monitoring abnormal changes of futures market spreads.<\/p>\n\n\n\nNew research papers related to existing strategies:<\/strong><\/h5>\n\n\n\n
Abstracto:
Brand buzz sentiment – the favorability of public communications about the brand – has become a major source to gauge brand reputation and predict investor returns. In this research, the authors build a prediction model of investor returns by empirically elaborating on its intricate relations with two novel sentiment measures by integrating brand reputation literature with brand buzz research. Accordingly, investor returns are conceptualized as a function of the favorability of buzz associated with the brand’s ability to deliver its outputs (brand ability buzz sentiment) and the favorability of buzz associated with the brand’s societal impact (brand responsibility buzz sentiment) along with their interaction, respectively. Deploying support vector machine learning and panel vector autoregression, preliminary evidence suggests that brand ability buzz sentiment but not brand responsibility buzz sentiment drives investor returns, yet their interaction inhibits investor returns. The proposed model outperforms extant prediction models of investor returns.<\/p>\n\n\n\n
https:\/\/ssrn.com\/abstract=4277189<\/a>
Abstracto:
Funds with an environmental, social and corporate governance (ESG) mandate have been growing rapidly in recent years and received in\ufb02ows also during periods of market turmoil, such as March 2020, in contrast to their non-ESG peers. This paper investigates whether investors in ESG funds react di\ufb00erently to past negative performance, making these funds less sensitive to short-term changes in returns. In the absence of an ESG-label, we de\ufb01ne an ESG- or Environmentally-focused fund if its name contains relevant words. The results show that ESG\/E equity and corporate bond funds exhibit a weaker \ufb02ow-performance relationship compared to traditional funds in 2016-2020. This \ufb01nding may re\ufb02ect the longer-term investment horizon of ESG investors and their expectation of better risk-adjusted performance from ESG funds in the future. We also explore how the results vary across institutional and retail investors and how they depend on the liquidity of funds\u2019 assets and wider market conditions. A weaker \ufb02ow-performance relationship allows funds to provide a stable source of \ufb01nancing to the green transition and may reduce risks for \ufb01nancial stability, particularly during turmoil episodes.<\/p>\n\n\n\n
https:\/\/ssrn.com\/abstract=4310211<\/a>
Abstracto:
Greenwashing is framed as a signalling game between firms and investors and occurs due to cheap news signals. A worst social outcome occurs with a pooling equilibrium. Firms greenwashed if ex-ante they are in the top decile of a portfolio sorted by ESG news scores from NLP, and ex-post fell out from the top decile for ESG performance scores. The greenwashing premium for ESG in USA is historically not significant but episodic. The premium in 2020 was largely due to consumer and green firms at 2.9% and 4.2% respectively. It was larger for the \u2018E\u2019 and \u2018G\u2019 than the \u2018S\u2019 pillar, except amongst brown firms. A mechanism design is proposed to conceptually incentivise firms to keep to their ESG goals.<\/p>\n\n\n\n
https:\/\/ssrn.com\/abstract=4346043<\/a>
Abstracto:
We test the out-of-sample trading performance of model-free reinforcement learning (RL) agents and compare them with the performance of equally-weighted portfolios and traditional mean-variance (MV) optimization benchmarks. By dividing European and U.S. indices constituents into factor datasets, the RL-generated portfolios face different scenarios defined by these factor environments. The RL approach is empirically evaluated based on a selection of measures and probabilistic assessments. Training these models only on price data and features constructed from these prices, the performance of the RL approach yields better risk-adjusted returns as well as probabilistic Sharpe ratios compared to MV specifications. However, this performance varies across factor environments. RL models partially uncover the nonlinear structure of the stochastic discount factor. It is further demonstrated that RL models are successful at reducing left-tail risks in out-of-sample settings. These results indicate that these models are indeed useful in portfolio management applications.<\/p>\n\n\n\n
<\/strong>https:\/\/ssrn.com\/abstract=4342478<\/a>
Abstracto:
This paper aims to enhance the classical mean-variance portfolio selection by using machine learning techniques and accounting for systemic risk. The optimal portfolio is solved through a three-step supervised learning model. Firstly, the Smooth Pinball Neural Network is employed to predict return distributions of individual assets and the market. Secondly, we use copula to model dependence between assets and the market, based on which we simulate return scenarios. Lastly, we maximize an ex-ante conditional Sharpe ratio conditioning on systemic events. We run a large-scale comparative study using nearly 600 US individual stocks over 37 years. Our set of predictors includes 94 firm characteristics, 14 macroeconomic variables, and 74 industry dummies. The backtesting results demonstrate the superiority of our proposed approach over popular benchmark strategies including a GARCH-based model. This outperformance is statistically significant and robust to the inclusion of transaction costs.<\/p>\n\n\n\n
https:\/\/ssrn.com\/abstract=4384871<\/a>
Abstracto:
In this article, we propose a novel approach to developing an end-to-end stock trading strategy using Deep Reinforcement Learning. Our model integrates a multi-view environment representation neural network and a Long Memory mechanism to enhance its ability to explore strategy and improve the utilization of simulated transaction generated during the learning process. Specifically, we classify 20 indicators into several categories and utilize them as inputs to our model. The multi-view environment representation network is designed to perceive the market and derive the trading strategy from the model output. Our empirical results shows that our strategy outperforms the market and …<\/p>\n\n\n\n
https:\/\/ssrn.com\/abstract=4347633<\/a>
Abstracto:
Using a comprehensive corporate governance data set and machine learning techniques, specifically Neural Networks, I validate a significant effect and out of sample predictability of governance for corporate bond yield to maturity, which is the most important discount factor for bond valuations. Institutional investors are aware of the issuers\u2019 governance profiles in making asset allocations, and governance also impacts the investors\u2019 portfolio returns. Among the institutional investors, the long term and dominant insurance companies in the corporate bond market are the most sensible in assessing the governance quality, and their portfolio returns correspond the most sensitively to governance.<\/p>\n\n\n\nAnd several interesting free blog posts have been published during last 2 weeks:<\/strong><\/h5>\n\n\n\n
Plus, the following trading strategies have been backtested in QuantConnect<\/a> in the previous two weeks:<\/strong><\/h4>\n\n\n\n
#842 – Cross-Sectional Mood Beta Strategy in Equities<\/a>
#867 – Cross Industry Dispersion Factor<\/a>
#868 – Industry-Relative Stock Beta in Chinese Equities<\/a><\/p>","protected":false},"excerpt":{"rendered":"