Quantpedia Premium Update – 16th November 2021

New strategies:

#684 –Volatility Arbitrage Based on Breakeven Volatility

Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: options, ETFs
Complexity: Very complex strategy
Backtest period: 2015-2020
Indicative performance: 6.8%
Estimated volatility: 4.8%

Source paper:

Hull, Blair and Li, Anlong and Qiao, Xiao: Option Pricing Via Breakeven Volatility
https://ssrn.com/abstract=3938897
Abstract:
The fair value of an option is given by breakeven volatility, the value of implied volatility that sets the profit and loss of a delta-hedged option to zero. We calculate breakeven volatility for 400,000 options traded on the S&P 500 Index, and we build a predictive model for these volatility values. A two-stage regression approach captures the majority of observed variation. By providing a link between option characteristics, such as moneyness and time to expiration, and breakeven volatility, we establish a nonparametric approach of pricing options without the need to specify the underlying price process. We illustrate the economic value of our approach with a simulated trading strategy based on model predictions of breakeven volatility.

#685 – Boosted Trees and Cryptocurrency Return Prediction

Period of rebalancing: Daily
Markets traded: cryptos
Instruments used for trading: cryptos
Complexity: Very complex strategy
Backtest period: 2010-2021
Indicative performance: 21.89%
Estimated volatility: 9.42%

Source paper:

Filippou, Ilias and Rapach, David and Rapach, David and Thimsen, Christoffer: Boosting Cryptocurrency Return Prediction
https://ssrn.com/abstract=3914414
Abstract:
We use boosted decision trees to generate daily out-of-sample forecasts of excess returns for Bitcoin and Ethereum, the two best-known and largest cryptocurrencies. The decision trees incorporate information from 39 predictors, including variables relating to cryptocurrency fun- damentals, technical indicators, Google Trends searches, Reddit comments, and articles from Factiva. We use the XGBoost algorithm to boost trees and find that excess return forecasts based on boosted trees produce statistically and economically significant out-of-sample gains. We explore the importance of individual predictors and nonlinearities in the fitted boosted trees. We find that a broad array of predictors are relevant for forecasting daily cryptocurrency returns and that strong nonlinearities characterize the predictive relationships.

#686 – Stock Issuance Effect

Period of rebalancing: Yearly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Simple strategy
Backtest period: 1963-2020
Indicative performance: 6.17%
Estimated volatility: 4.29%

Source paper:

Tim Loughran, Reconsidering Equity Issue Performance: A Focused Criticism of the Fama-French Factor Models (October 22, 2021)
https://ssrn.com/abstract=3907523
Abstract:
The Fama and French (2015) 5-factor model is commonly used to measure the performance of stock return portfolios. Importantly, we find that three of the Fama and French (2015) firm-level characteristics (i.e., size, BV/MV, and profitability) have no significant explanatory power in the cross-section of returns for companies above the median NYSE capitalization during 1963-2020. Small firms comprising less than 8% of the total market capitalization drive the patterns of the 5-factor model. This paper also reexamines equity issuer performance in the context of the 5-factor firm level characteristics and finds that small and large issuers have similar underperformance.

#687 –Intangible Factor in US Equities

Period of rebalancing: Yearly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Complex strategy
Backtest period: 1989-2020
Indicative performance: 4.66%
Estimated volatility: 8.87%

Source paper:

Bongaerts, Dion and Kang, Xiaowei and Van Dijk, Mathijs A., Intangible assets and the cross-section of stock returns
https://ssrn.com/abstract=3927990
Abstract:
We examine whether intangible assets are priced in the cross-section of stock returns. We find that intangible asset intensity has more explanatory power than size, value, profitability, and investment. An intangibles-based long-short factor has a higher Sharpe ratio than these established factors. Adding the intangible factor to the Fama-French five-factor model improves the description of average returns and makes the investment factor redundant. The intangible factor is distinct from traditional growth strategies, provides a hedge to value and quality strategies, and expands investors’ opportunity sets. Intangible intensity as characteristic is more important than as risk factor, consistent with intangibles-based mispricing.

#688 – Relative Value Factor in US

Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Complex strategy
Backtest period: 1976-2019
Indicative performance: 11%
Estimated volatility: 9.32%

Source paper:

Hu, Xiaolu and Sy, Malick O. and Sy, Malick O. and Wu, Liuren, A Factor Model of Company Relative Valuation
https://ssrn.com/abstract=3706995
Abstract:
Accurate company valuation is the starting point of value investing and corporate decisions. This paper proposes a statistical factor model to generate company valuation comparison across a large universe. The model scales the market value of a company by its book capital to generate a cross-sectionally comparable relative value target, constructs valuation factors by combining several descriptors from a similar category to increase coverage and reduce multicollinearity, and links industry classification and the valuation factors to the company relative value via a cross-sectional contemporaneous regression at each date. Historical analysis on U.S. publicly traded companies shows that the factor model explains a large proportion of the cross-sectional variation of company relative value and experiences little out-of-sample degeneration. The regression residual represents temporary company misvaluation, and can be exploited by both outside investors as attractive investment opportunities and internal management for market timing of corporate decisions.

New research papers related to existing strategies:

#283 – Market Neutral Strategy Based on Share Buybacks
#138 – Repurchase/New issue Effect

Rehman, Obaidur: The Price Impact and Timing of Actual Share Repurchases in Norway
https://ssrn.com/abstract=3796599
Abstract:
Little is known about the price impact and timing of actual share repurchases. Data unavailability has hindered research in most countries, including the United States. Using unique data on actual share repurchase transactions from Norway, we test for the price impact and timing of daily open market repurchases. We find evidence that share repurchases typically follow after a negative drift in the stock price, and the average three-day abnormal return around the announcement is 0.54%. Moreover, the initial market reaction is greater for repurchases that are pursued by small firms and for firms that experience a negative drift in the stock price prior to the transaction. The evidence presented is seemingly indicative of managers’ intent to signal undervaluation through repurchase transactions. However, we do not find any significant long-term abnormal returns for repurchasing firms. This result suggests that on average, managers do not time the market based on informational advantage.

#35 – Insiders Trading Effect in Stocks

Avci, Sureyya Burcu: Insider Trading Profitability in Turkey
https://ssrn.com/abstract=3896937
Abstract:
This study presents the first large-scale, comprehensive evidence on the insider trading patterns and abnormal returns in the Turkish stock market. Starting with a summary of the legislation, an event study methodology is used to compute daily abnormal returns of almost 65,000 insider transactions. Findings show that insiders earn 1.56 percent more than the market average return on the first six days following the trading day. The highest abnormal returns are earned by small firm insiders. Top executives, officers, directors, legal entities, funds, and large shareholders earn significantly higher than the market average return. Short-term and midterm abnormal profits vary with size, the value of the trade, holdings of the insider, relation of the insider with the company, number of insiders within a company, and whether the transaction is a sale or purchase.

#202 – Dividend Announcement Effect

Groenke, Robert: Dividend Redux
https://ssrn.com/abstract=3837937
Abstract:
Has dividend equity investing as a style been rendered irrelevant by the recent rise of equities without a payout, or should it remain an important building block in modern portfolios? This paper is an attempt to explore this question, supported by nearly a century of data. Reviewing the history of returns, I find strong evidence demonstrating long-term risk-adjusted return advantages for dividend equities relative to non-payers and the market despite a recent flourish from non-payers. I also find that despite the higher realized average returns and Sharpe ratios for a representative mid-yield dividend category, higher volatility and the relative propensity and size of right-tail outcomes for the market and non-dividend equities, in particular, make periods of relative underperformance not only possible but reasonably probable.

#615 – Machine Learning and Mutual Fund Characteristics

DeMiguel, Victor and Gil-Bazo, Javier and Nogales, Francisco J. and A. P. Santos, Andre: Can Machine Learning Help to Select Portfolios of Mutual Funds?
https://ssrn.com/abstract=3768753
Abstract:
Identifying outperforming mutual funds ex-ante is a notoriously difficult task. We use machine learning to exploit fund characteristics and construct portfolios of equity funds that earn positive and significant out-of-sample alpha net of all costs. In contrast, alphas of portfolios selected with OLS are indistinguishable from zero. We show that the performance of machine-learning methods is the joint outcome of exploiting multiple fund characteristics and allowing for flexibility in the relation between characteristics and performance. Our results hold also for portfolios of only retail funds, for various measures of fund performance, for different methodological choices, and across different market conditions.

#57 – Term Spread Premium

Choi, Jaehyuk and Ge, Desheng and Kang, Kyu H. and Sohn, Sungbin: Predicting Recession Probabilities Using Term Spreads: New Evidence from a Machine Learning Approach
https://ssrn.com/abstract=3723717
Abstract:
The literature on using yield curves to forecast recessions typically measures the term spread as the difference between the 10-year and the three-month Treasury rates. Furthermore, using the term spread constrains the long- and short-term interest rates to have the same absolute effect on the recession probability. In this study, we adopt a machine learning method to investigate whether the predictive ability of interest rates can be improved. The machine learning algorithm identifies the best maturity pair, separating the effects of interest rates from those of the term spread. Our comprehensive empirical exercise shows that, despite the likelihood gain, the machine learning approach does not significantly improve the predictive accuracy, owing to the estimation error. Our finding supports the conventional use of the 10-year–three-month Treasury yield spread. This is robust to the forecasting horizon, control variable, sample period, and oversampling of the recession observations.

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

What Drives Volatility of Bitcoin?

Extremely high bitcoin returns and drawdowns come hand in hand with significant volatility. As Bitcoin is becoming an unignorable part of finance with substantial institutional participation, it is necessary to understand the key drivers of returns and volatility, which is comparably persistent as in other, more established asset classes. In addition, other cryptocurrencies are extremely correlated with Bitcoin, so understanding of key drivers of Bitcoin volatility might also carry to other cryptos. The research of Lyócsa et al. (2020) examines several possible drivers of the volatility. The authors study the realized volatility and its jump component and identify whether the volatility is influenced by various factors such as news about the regulation of bitcoin, hacking attacks on bitcoin exchanges, investor sentiment, and various types of macroeconomic news. The study identifies the significant impact of two intuitive factors: news about the regulation or cryptocurrency exchange hacks. Lagged volatility is also an essential factor, as shown by regression analysis. Regarding macroeconomic data, economic fundamentals do not seem to influence the volatility, except for forward-looking indicators (e.g., the consumer confidence index). Lastly, the authors study the investor sentiment extracted from Google searches, but only the positive sentiment has some impact. Overall, the research is a vital addition to the literature that helps us understand Bitcoin’s volatility.

Bitcoin Returns and Volatility Predicted by Bitcoin Exchange Reserves

In the modern world full of technologies, cryptocurrencies are gaining popularity every day. The most famous cryptocurrency, bitcoin, was introduced in 2009. Ever since its launch and its subsequent success, when within a few years, its price skyrocketed, and it has been the subject of many price predicting studies. These, however, primarily focus on the market and macro factors, entirely omitting the nature of bitcoin – which is blockchain technology. In this study, authors Hoang and Baur try to capture and research this interconnection between behaviour of investors, bitcoin exchanges, and blockchain.

How News Move Markets ?

Nobody would argue that nowadays, we live in an information-rich society – the amount of available information (data) is constantly rising, and news is becoming more accessible and frequent. It is indisputable that this evolvement has also affected financial markets. Machine learning algorithms can chew up big chunks of data. We can analyze the sentiment (which is frequently related to the news). Big data does not seem to be a problem anymore, and high-frequent trading algorithms can react almost instantly. But how important is the news? Kerssenfischer and Schmeling (2021) provide several answers by studying the impact of scheduled and unscheduled news (frequently omitted in other news-related studies) in connection with high-frequency changes in bond yields and stock prices in the EU and US as well. The research points out that the effect is tremendous and significant.

How to Combine Different Momentum Strategies

Today we will again talk more about the portfolio management theory, and we will focus on techniques for combining quantitative strategies into one multi-strategy portfolio. So, let’s imagine we already have a set of profitable investment strategies, and we need to combine them. The goal of such “strategy allocation” usually is to achieve the best risk-adjusted return possible. There is no single correct solution to this task, but there are a few methods that we can try. The “appropriate combination” highly depends on the type of strategies we are about to combine. Are we combining equity and bond strategies together? Are we combining equity strategies, with each one having an entirely different logic? Or do we rather need to assign weights to strategies that are similar in nature yet still different? We will focus this article on the last option – combining similar yet different strategies.