New strategies:
#535 – Idiosyncratic Liquidity in Stocks
Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Moderately complex strategy
Backtest period: 1966-2018
Indicative performance: 11.09%
Estimated volatility: 11.3%
Source paper:
Baris Ince: Systematic vs. Idiosyncratic Liquidity: Cross-section of Stock Returns
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3644447
Abstract:
This paper decomposes firm-specific monthly-varying Amihud (2002) illiquidity measure into two components: (i) systematic illiquidity; (ii) idiosyncratic illiquidity. While there is a positive and significant relationship between systematic illiquidity and one-month-ahead stock returns, the observed relationship disappears when very small stocks are excluded. On the other hand, investors tend to underreact to idiosyncratic (il)liquidity. Hence, stocks with positive (negative) idiosyncratic liquidity generate positive (negative) subsequent returns. More specifically, high-low idiosyncratic liquidity strategy generates around 11% annualized value-weighted risk-adjusted return. Investor inattention, illiquidity, and sentiment-driven irrational investors are the main drivers of underreaction to idiosyncratic liquidity component.
#536 – Machine Learning Stock Picking
Period of rebalancing: Weekly
Markets traded: equities
Instruments used for trading: CFDs, stocks
Complexity: Very complex strategy
Backtest period: 2002-2019
Indicative performance: 18.7%
Estimated volatility: 24.2%
Source paper:
Wolff, Dominik and Echterling, Fabian: Stock Picking with Machine Learning
https://ssrn.com/abstract=3607845
Abstract:
We combine insights from machine learning and finance research to build machine learning algorithms for stock selection. Our study builds on weekly data for the historical constituents of the S&P 500 over the period from 1999 to 2019 and includes typical equity factors as well as additional fundamental data, technical indicators, and historical returns. Deep Neural Networks (DNN), Long Short-Term Neural Networks (LSTM), Random Forest, Boosting, and Regularized Logistic Regression models are trained on stock characteristics to predict whether a specific stock outperforms the market over the subsequent week. We analyze a trading strategy that picks stocks with the highest probability predictions to outperform the market. Our empirical results show a substantial and significant outperformance of machine learning based stock selection models compared to a simple equally weighted benchmark. Moreover, we find non-linear machine learning models such as neural networks and tree-based models to outperform more simple regularized logistic regression approaches. The results are robust when applied to the STOXX Europe 600 as alternative asset universe. However, all analyzed machine learning strategies demonstrate a substantial portfolio turnover and transaction costs have to be marginal to capitalize on the strategies.
#537 – The Positive Similarity of Company Filings and Stock Returns
Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Moderately complex strategy
Backtest period: 2007-2020
Indicative performance: 5.47%
Estimated volatility: 6.48%
Source paper:
Padyšák, Matúš: The positive similarity of company filings and the cross-section of stock returns
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3690461
Abstract:
It is already well-documented that textual analysis of 10-K & 10-Qs can be largely profitable. This research studies the similarity of language used in the filings using data which enables to analyze what type of language is similar. Results show that the similarity of the positive language is the most profitable option. From a practical point of view, the positive similarity effect is examined. Results show that the lowest positive similarity stocks significantly outperform the highest positive similarity stocks. The effect cannot be explained by the common asset pricing models, nor by the change of sentiment in the financial reports. Therefore, the positive similarity effect could be considered as a distinct anomaly in the financial markets. In the long-only implementation, the strategy is highly profitable, and in the long-short implementation, the strategy has impressive consistency and risk-adjusted return (0.84).
#538 – Oil Intraday Momentum
Period of rebalancing: Intraday
Markets traded: commodities
Instruments used for trading: futures, ETFs, CFDs
Complexity: Simple strategy
Backtest period: 2009-2020
Indicative performance: 21.36%
Estimated volatility: not stated
Source paper:
Guglielmo Maria Caporale, Alex Plastun: Gold and Oil Prices: Abnormal Returns, Momentum and Contrarian Effects
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3662052
Abstract:
his paper explores price (momentum and contrarian) effects on the days characterised by abnormal returns and the following ones in two commodity markets. Specifically, using daily Gold and Oil price data over the period 01.01.2009-31.03.2020 the following hypotheses are tested: H1) there are price effects on days with abnormal returns, H2) there are price effects on the day after abnormal returns occur; H3) the price effects caused by abnormal returns are exploitable. For these purposes average analysis, t-tests, CAR and trading simulation approaches are used. The main results can be summarised as follows. Hourly returns during the day of abnormal returns are significantly bigger than those during average “normal” days. Prices tend to move in the direction of abnormal returns till the end of the day when these occur. The presence of abnormal returns can usually be detected before the end of the day by estimating specific timing parameters, and a momentum effect can be detected. On the following day two different price patterns are detected: a momentum effect for Oil prices and a contrarian effect for Gold prices respectively. Trading simulations show that these effects can be exploited to generate abnormal profits.
New research papers related to existing strategies:
#460 – ESG Level Factor Investing Strategy
Conen, Ralf, Hartmann, Stefan and Rudolf, Markus: Going Green Means Being in the Black
https://ht323.infusion-links.com/api/v1/click/5991057051484160/6713207214571520
Abstract:
We revisit the relationship between ESG and stock returns using a novel, monthly consensus rating for a global universe. Our results illustrate that how well or bad a firm does along the different dimensions of corporate social responsibility does affect the return of its shares. Fund managers constructing portfolios using information on a firms’ corporate social performance generally outperform, however may underperform in markets, where social responsibility is not as widely accepted. Secondly, excess performance of portfolios tilted towards corporate social responsibility is not always fully explained by the interaction with common risk factors such as value, size or momentum suggesting that ESG has a systematic effect on stock returns beyond those factors. This enables active fund managers to harvest risk-adjusted alpha. Thirdly, the effect of ESG on portfolio performance is asymmetric and does not appear to be constant over time. Fourth, markets reward short and longterm performance along ESG dimensions differently. Lastly, ESG is not a globally integrated factor. Rather it differs across regions with regard to direction, magnitude and statistical significance. We do not find a scenario in which investing in stocks with high ESG ratings leads to negative risk adjusted performance, suggesting that investors can greenwash portfolios without sacrificing performance.
#25 – Size Factor – Small Capitalization Stocks Premium
Blitz, David and Hanauer, Matthias Xaver, Settling the Size Matter
https://ssrn.com/abstract=3686583
Abstract:
The size premium has failed to materialize since its discovery almost forty years ago, but is seemingly revived when controlling for quality-versus-junk exposures. This paper aims to resolve whether there exists a distinct size premium that can be captured in reality. For the US we confirm that a highly significant alpha emerges in regressions of size on quality, but for international markets we find that the size premium remains statistically indistinguishable from zero. Moreover, the US size premium appears to be beyond the practical reach of investors, because the alpha that is observed ex post in regressions cannot be captured by controlling for quality exposures ex ante. We also find that the significant regression alpha in the US is entirely driven by the short side of quality. Altogether, these results imply that size only adds value in conjunction with a short position in US junk stocks. However, we also show that small-cap exposure is vital for unlocking the full potential of other factors, such as value and momentum. We conclude that size is weak as a stand-alone factor but a powerful catalyst for other factors.
#453 – Machine Learning Adaptive Portfolio Asset Allocation
Zhang, Zihao and Zohren, Stefan and Roberts, Stephen: Deep Learning for Portfolio Optimisation
https://ssrn.com/abstract=3613600
Abstract:
We adopt deep learning models to directly optimize the portfolio Sharpe ratio. The framework we present circumvents the requirements for forecasting expected returns and allows us to directly optimize portfolio weights by updating model parameters. Instead of selecting individual assets, we trade Exchange-Traded Funds (ETFs) of market indices to form a portfolio. Indices of different asset classes show robust correlations and trading them substantially reduces the spectrum of available assets to choose from. We compare our method with a wide range of algorithms with results showing that our model obtains the best performance over the testing period, from 2011 to the end of April 2020, including the financial instabilities of the first quarter of 2020. A sensitivity analysis is included to understand the relevance of input features and we further study the performance of our approach under different cost rates and different risk levels via volatility scaling.
#530 – Jump Risk in Stocks
Zambon, Nancy and Caporin, Massimiliano and Distaso, Walter: Jump risk and pricing implications
http://wp.lancs.ac.uk/fofi2018/files/2018/03/FoFI-2018-0102-Walter-Distaso.pdf
Abstract:
This paper identifies a new common risk factor in stock returns related to the fear of future jumps: the Jump Factor. It is possible to include the factor in standard asset-pricing models leading to a fivefactor model which is directed at capturing the size, value, profitability, momentum and jump expectation in stock returns. Standard analysis show that the jump component proxy, in stock returns, for sensitivity to a common risk factor and that the Jump Factor is able to explain much of the variation in returns both in time and cross-section. Moreover, the risk premia associated with the Jump Factor is negative, significant, and close to its factor portfolio mean excess returns.
#27 – Market Timing with Aggregate and Idiosyncratic Stock Volatilities
Detzel, Andrew L. and Duarte, Jefferson and Kamara, Avraham and Siegel, Stephan and Sun, Celine: The Cross-Section of Volatility and Expected Returns: Then and Now
https://ssrn.com/abstract=3455609
Abstract:
We successfully replicate the main results of Ang, Hodrick, Xing, and Zhang (2006): Aggregate-volatility risk and idiosyncratic volatility (IV) are each priced in the cross-section of stock returns from 1963 to 2000. We also examine the pricing of volatility outside the original time period and under more recent asset-pricing models. With the exception of NASDAQ stocks, aggregate-volatility risk continues to be priced in the years following the Ang et al. (2006) sample period, and none of the more recent asset-pricing models we consider consistently accounts for the pricing of aggregate-volatility risk. The difference in abnormal returns between stocks with high and low IV decreases but remains significant out-of-sample. More recent asset-pricing models do not resolve the IV anomaly for the Ang et al. (2006) sample, but the four-factor model of Stambaugh and Yuan (2017) and the six-factor model of Barillas and Shanken (2018) resolve the anomaly out-of-sample and over the extended period of 1967 to 2016. Finally, both models eliminate the arbitrage asymmetry that Stambaugh, Yu, and Yuan (2015) propose as an explanation of the IV anomaly.
#383 – Moving Average Strategies for Cryptocurrencies
Detzel, Liu, Strauss, Zhou, Zhu: Bitcoin: Predictability and Profitability via Technical Analysis
https://www.paris-december.eu/sites/default/files/pdf/parismeeting/2018/STRAUSS_2018.pdf
Abstract:
We document that Bitcoin returns, while unpredictable by macroeconomic variables, are predictable by 1- to 20-week moving averages (MAs) of daily prices, both in- and out-of-sample. Trading strategies based on MAs generate substantial alpha, utility and Sharpe ratios gains, and signi cantly reduce the severity of drawdowns relative to a buyand- hold position in Bitcoin, which already has a Sharpe ratio of 1.8. We explain these facts with a novel equilibrium model that demonstrates, with uncertainty about growth in fundamentals, rational learning by investors with di erent priors yields predictability of returns by MAs.
And three interesting free blog posts have been published during last 2 weeks:
The Positive Similarity of Company Filings and the Cross-Section of Stock Returns
The usage of alternative data is now a main-stream topic in investment management and algorithmic trading. So let’s together explore the textual analysis of 10-K & 10-Q filings and analyze how these datasets could be used as a profitable part of investment portfolios. We invite you to read this short summarization of the research. Full version can be found on the SSRN.
Equity factors are not as straightforward as they may seem to be. There is an ongoing debate about their usability or performance since they have a cyclical performance. Moreover, the modern trend of smart beta only fuels this debate. Novel research by Blitz and Hanauer examines the size factor. While many argue that the size premium does not exist and the size factor is useless, this paper suggests that the size factor can be an important addition to the other factors. To be more precise, the size factor can be an important addition to the quality, value or momentum factors.
Exchange-traded funds (ETFs) have become popular and important investment vehicles in the financial markets. However, that is not a shock given the numerous benefits connected with ETFs. Naturally, they have caught the interest of academics, and there is plenty of literature about ETFs. While the profits and trading strategies are probably the most important research topics for practitioners, liquidity in the financial markets is almost equally important. Concerning liquidity in the ETFs, novel research by Pham et al. shows when exactly are ETFs the most liquid. Looking on the spreads, they are the lowest at market close. Such a finding can be an essential part of an optimal trading position making, where the aim is to minimize the trading costs.
Plus, the following six trading strategies have been backtested in QuantConnect in the previous two weeks:
#370 – Value-Growth Timing
#512 – Savings Rate Beta and Performance of Stocks
#520 – Lottery Stocks and Past Performance
#521 – Global Dollar Risk Strategy
#531 – Global Bond Portfolio Predicts Government Bonds Returns in Individual Countries
#533 – FOMC Cycle and Credit Risk
Are you looking for more strategies to read about? Sign up for our newsletter or visit our Blog or Screener.
Do you want to learn more about Quantpedia Premium service? Check how Quantpedia works, our mission and Premium pricing offer.
Do you want to learn more about Quantpedia Pro service? Check its description, watch videos, review reporting capabilities and visit our pricing offer.
Are you looking for historical data or backtesting platforms? Check our list of Algo Trading Discounts.
Would you like free access to our services? Then, open an account with Lightspeed and enjoy one year of Quantpedia Premium at no cost.
Or follow us on:
Facebook Group, Facebook Page, Twitter, Linkedin, Medium or Youtube
