Quantpedia Premium Update – 15th February 2021

New strategies:

#591 – Machine Learning and Earnings Announcements

Period of rebalancing: Daily
Markets traded: equities
Instruments used for trading: stocks
Complexity: Very complex strategy
Backtest period: 2013-2019
Indicative performance: 11.63%
Estimated volatility: 8.16%

Source paper:

Schnaubelt, Matthias and Seifert, Oleg: Valuation ratios, surprises, uncertainty or sentiment: How does financial machine learning predict returns from earnings announcements?
https://www.iwf.rw.fau.de/files/2020/04/04_2020.pdf
Abstract:
We apply state-of-the-art financial machine learning to assess the return-predictive value of more than 45,000 earnings announcements on a majority of S&P1500 constituents. To represent the diverse information content of earnings announcements, we generate predictor variables based on various sources such as analyst forecasts, earnings press releases and analyst conference call transcripts. We sort announcements into decile portfolios based on the model’s abnormal return prediction. In comparison to three benchmark models, we find that random forests yield superior abnormal returns which tend to increase with the forecast horizon for up to 60 days after the announcement. We subject the model’s learning and out-of-sample performance to further analysis. First, we find larger abnormal returns for small-cap stocks and a delayed return drift for growth stocks. Second, while revenue and earnings surprises are the main predictors for the contemporary reaction, we find that a larger range of variables, mostly fundamental ratios and forecast errors, is used to predict post-announcement returns. Third, we analyze variable contributions and find the model to recover non-linear patterns of common capital markets effects such as the value premium.
Leveraging the model’s predictions in a zero-investment trading strategy yields annualized returns of 11.63 percent at a Sharpe ratio of 1.39 after transaction costs.

#592 – Volatility effect in the Chinese A-share market

Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Simple strategy
Backtest period: 2000-2018
Indicative performance: 10.06%
Estimated volatility: 21.2%

Source paper:

David Blitz, Matthias X. Hanauer and Pim van Vliet : The Volatility Effect in China
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3765625
Abstract:
This paper shows that low-risk stocks significantly outperform high-risk stocks in the local China A shares market. The main driver of this low-risk anomaly is volatility, and not beta. A Fama-French style VOL factor is not explained by the Fama-French-Carhart factors, and has the strongest stand-alone performance among all these factors. Our findings are robust across sectors and over time, and consistent with previous empirical evidence for the US and international markets. Moreover, the VOL premium exhibits excellent investability characteristics, as it involves a low turnover and remains strong when applied to only the largest and most liquid stocks. Our results imply that the volatility effect is a highly pervasive phenomenon, and that explanations should be able to account for its presence in highly institutionalized markets, such as the US, but also in the Chinese market where private investors dominate trading.

#593 – Value in Cryptocurrencies

Period of rebalancing: Weekly
Markets traded: cryptos
Instruments used for trading: cryptos
Complexity: Simple strategy
Backtest period: 2017-2020
Indicative performance: 130.63%
Estimated volatility: 72.51%

Source paper:

Luca Liebi: Is There a Value Premium in Cryptoasset Markets?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3718684
Abstract:
This paper identifies active addresses-to-network value as an additional common risk factor in the returns on cryptoassets. Active addresses refer to the number of unique wallet addresses that conduct an on-chain transaction, whereas the network value of a cryptoasset corresponds to its market capitalization. Investigating 652 cryptoassets, I find that there are anomalous returns that increase with active addresses-to-network value ratio, a proxy for the value anomaly. Cryptoassets with a high active address to network value ratio yield on average 2.1 percentage points higher weekly returns compared to cryptoassets with low active addresses to network value ratio, and comparable size. A four-factor model directed at capturing the value pattern in average returns performs better than a three-factor model, including the market, size, and momentum factor. Importantly, the results suggest that cryptoasset prices are related to their fundamentals.

#594 – Size in Cryptocurrencies

Source paper:

#595 – Notional Value Effect in Futures Markets

Period of rebalancing: Daily
Markets traded: equities, bonds, currencies, commodities
Instruments used for trading: futures
Complexity: Moderately complex strategy
Backtest period: 1990-2019
Indicative performance: 3.38%
Estimated volatility: 4.74%

Source paper:

Theodosios Athanasiadis: Notional Value Effect in Futures Markets
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3725085
Abstract:
We examine how the notional value of futures contracts predicts the cross-section of returns within the major asset classes tracking a large number of futures contracts. We find that low notional value contracts outperform high notional value contracts within government bonds, short-term rates, commodities, currencies, and equity indexes. A diversified portfolio of the strategy delivers abnormal returns after controlling for standard asset pricing factors. The strategy is related to value and reversal factors but their explanatory power is low. Differences in liquidity explain a large portion of the cross-section of notional value, where high notional value contracts are more liquid, and subsume the reversal and value factors. Volatility risk can be a partial explanation both cross-sectionally where low notional value contracts exhibit higher volatility and across time where shocks to market volatility decrease strategy returns.

New research papers related to existing strategies:

#455 – Nonlinear Support Vector Machines and Stock Picking
#536 – Machine Learning Stock Picking

Ersan, Deniz; Nishioka, Chifumi; Scherp, Ansgar: Comparison of machine learning methods for financial time series forecasting at the examples of over 10 years of daily and hourly data of DAX 30 and S&P 500
https://repository.kulib.kyoto-u.ac.jp/dspace/bitstream/2433/250454/1/s42001-019-00057-5.pdf
Abstract:
This article conducts a systematic comparison of three methods for predicting the direction (+/-) of financial time series using over ten years of DAX 30 and the S&P 500 datasets at daily and hourly frames. We choose the methods from representative machine learning families, particularly supervised versus unsupervised. The methods are Support Vector Machines (SVM), Artificial Neural Networks, and k-Nearest Neighbor (k-NN). We explore the influence of different training window lengths and numbers of out-of-sample predictions. Furthermore, we investigate whether Kernel Principle Component Analysis (KPCA) improves prediction, through reducing data dimensionality. Additionally, we verify whether combining machine learning methods by Bootstrap Aggregating outperforms single methods. Key insights from the experiment are: All machine learning methods are in principle useful to predict the direction of (+/-) financial time series. But to our surprise, increasing the window size only helps to a certain extent for hou ly data, before it actually reduces the performance. The number of out-of-sample predictions had a small impact, while KPCA made a strong difference for SVM and k-NN. Finally, backtesting selected machines with a trading system on daily data revealed the the lazy learner k-NN outperforms the supervised approaches.

#75 – Federal Open Market Committee Meeting Effect in Stocks

Zhu, Xingyu, Volume Dynamics around FOMC Announcements
https://ssrn.com/abstract=3730543
Abstract:
The stock market volume decreases in anticipation of FOMC announcements and increases afterward. I find, in the cross-section, that stocks with higher market risk exposure experience greater volume changes. I also find that volume dynamics around FOMC announcements are unlikely to be attributable to changes in volatility. Instead, they are linked to discretionary liquidity trading resulting from the presence of private information. I set up a model that guides my investigation of the information environment in the stock market around FOMC announcements. Consistent with the model’s implication, volume dynamics are accompanied by changes in the information environment. I find that information asymmetry increases ahead of FOMC announcements, but only for high-beta stocks.

#224 – Profitability Factor Combined with Value Factor

Wahal, Sunil and Repetto, Eduardo, The Joint Distribution of Value and Profitability: International Evidence
https://ssrn.com/abstract=3739571
Abstract:
We examine the joint distribution of value and profitability in international markets. Over the 1990-2020 sample period, portfolios formed at the intersection of value and profitability generate systematic return patterns: value and high-profitability portfolios substantially outperform growth and low- profitability portfolios. The return differences are present in developed markets (ex-US), emerging markets, subsets and combinations thereof (EAFE, ACWI ex-US), as well as in large and small stocks. When viewed in combination with similar such evidence from the US, the data suggest important benefits to targeting value and profitability jointly rather than piecemeal.

#353 – US Sector Rotation with Five-Factor Fama-French Alphas

Umutlu, Mehmet and Bengitoz, Pelin, The Cross-section of Industry Equity Returns and Global Tactical Asset Allocation across Regions and Industries
https://ssrn.com/abstract=3720543
Abstract:
This study investigates the existence of a cross-sectional relationship between several index characteristics and future returns on country-industry indexes in six regions. The results from portfolio sorts and index-level cross-sectional regressions across regions collectively show that the geographical origin and size of indexes critically determine the existence of such a relationship. We find that country-industry indexes of any size with high earnings-to-price ratio yield higher expected returns in the US, Europe, and Asia-Pacific. Recent winner (loser) small-cap portfolios in North America and Asia-Pacific have a tendency to outperform (underperform) recent loser (winner) portfolios in the near future, while this kind of performance continuity is pervasive across all size groups of European portfolios. Small portfolios with high idiosyncratic volatility in Europe, Asia-Pacific, and Japan earn an idiosyncratic volatility premium. Dividend yield and short-term momentum are positively related to future returns of small European portfolios. Our results have implications for portfolio managers following a global tactical asset allocation policy.

#460 – ESG Level Factor Investing Strategy
#461 – ESG Factor Momentum Strategy

Abhayawansa, Subhash and Tyagi, Shailesh, Sustainable Investing: The Black Box of Environmental, Social and Governance (ESG) Ratings
https://ssrn.com/abstract=3777674
Abstract:
Environmental, social and governance (ESG) investing is becoming mainstream, and the COVID-19 pandemic has amplified the momentum. The interest in ESG investing has created greater demand for ESG data, ratings and rankings together with a proliferation of agencies offering these products which are unquestioningly relied on by investors, academics and regulators. Research highlights that different ESG ratings and rankings produce significantly different assessments of the ESG performance of companies. In this paper, we examine the causes of the differences in the ratings and ranking produced by different agencies. It is found that the divergences between raters can be attributed to differences in defining ESG constructs (i.e., theorisation problem) and methodological differences (i.e., commensurability problem). While users of ESG ratings and rankings are advised to study the definitions and methodologies prior to their use, lack of transparency about the data sources, weightings and methodologies makes it dificult to ensure that companies’ true ESG performance is accounted for when making portfolio selection and investment decisions. As a solution, we suggest that instead of attempting to compare and contrast ratings and rankings of different agencies, investors should determine ESG constructs material to their investment strategy and match them with an ESG ratings/rankings product that closely resemble those constructs.

And two interesting free blog posts have been published during last 2 weeks:

Basic Properties of Various Real Asset Portfolios

Do not put all your eggs in one basket is a common phrase that resonates among investors worldwide. The errand of such a famous saying is simple, diversify! However, how to diversify, if in the crisis, everything seems to be highly correlated? Last week, we wrote a blog about the Macro Factor Risk Parity, but it certainly is not the only option. Real assets such as REITs, various commodities, and the ever-popular gold are commonly added into portfolios as diversifiers. However, Parikh and Zhan (2019) research examine a much bigger set of real assets than the aforementioned evergreens. Real assets like Timberland, Farmland, Infrastructure, Natural Resources and many others are presented in the paper. All those assets have different sensitivities to inflation, GDP growth, equities or bonds. Therefore, real assets could have a value in the portfolios to protect an investor from inflation, stagnation, or simply distributing the eggs mentioned above in many baskets. All these strategies are presented in the paper and compared to equities, bonds and traditional 60/40.

Fake Trading on Crypto Exchanges

At Quantpedia, we acknowledge that cryptocurrencies offer numerous trading opportunities and include them in the Screener. Yet, each participant should be cautious. Cryptocurrencies are not black or white; they have their pros but also cons. Perhaps now, with all the positive sentiment around cryptos, it is the right time to advert also the cons. It is not that long time ago when we published a blog about the Bitcoin’s price manipulation, where the anecdotal evidence was supported by the Benford´s law which is related to the distribution of leading digits.

The novel research of Amiram et al. (2020) expands the previous work about the manipulation of the BTC. The authors include a tremendous amount of currencies, study various exchanges, and most importantly, they use more methods to examine the manipulations. To be more precise, the authors utilize the Benford´s law, deviations from the log-normal distribution and the novel machine-learning algorithm E-Divisive with medians that identifies structural breaks in time series. Moreover, they aggregate the measures by computing their principal components. While the results are as always best shown by the included figures, there are numerous practical suggestions. The fake trading benefits exchanges in the short term; however, it is harmful in the long term. Lastly, exchanges with the highest popularity, some regulations and the oldest ones tend to have the lowest fake trading levels.

Plus, the following six trading strategies have been backtested in QuantConnect in the previous two weeks:

#28 – Value and Momentum Factors across Asset Classes
#552 – Betting against beta in China
#554 – Size factor in China
#555 – Value factor in China
#557 – Profitability Factor in Chinese Equities
#568 – Momentum effect in Chinese B-shares

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