New strategies:
#606 – Climate Change Exposure and the Cross Section of Stock Returns
Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Moderately complex strategy
Backtest period: 1970-2019
Indicative performance: 5.13%
Estimated volatility: 15.89%
Source paper:
Xu, Jiangmin and Sun, Cheng and You, Yihui: Climate Change Exposure and Stock Return Predictability
https://ssrn.com/abstract=3777060
Abstract:
This paper finds evidence that stock returns vary with the climate change exposure of firms in a predictable manner. Using the Palmer Drought Severity Index, we construct firm-level climate change exposure and find that firms with high climate change exposure experience lower future profitability. We show that stock prices do not promptly incorporate such climate change information, and these firms with high climate change exposures are subject to subsequent lower stock returns. A long-short trading strategy based on this effect produces significant alphas of around 0.7% per month. Additionally, we find this return predictability is more pronounced under acute climate change conditions, and is robust to industry or macroeconomic factors.
#607 – Equity Momentum Spillover to Currencies
Period of rebalancing: Weekly
Markets traded: currencies
Instruments used for trading: futures, forwards, CFDs
Complexity: Moderately complex strategy
Backtest period: 1994 -2020
Indicative performance: 2.2%
Estimated volatility: 7.1%
Source paper:
Philippe Declerck: Do Equities Spill Over to Currencies?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3742800
Abstract:
We document that equities indices spill over to currencies: cross-sectional momentum signals based on equities returns can help building investment strategies in the currencies space. Like momentum, this spillover effect tends to works better for short / mid term lookback periods, but spillover does not seem to be only a momentum phenomenon. Spillover is also robust to signals and portfolio construction modifications.
#608 – Intraday Reversal in China
Period of rebalancing: Intraday
Markets traded: equities
Instruments used for trading: stocks
Complexity: Simple strategy
Backtest period: 2000-2019
Indicative performance: 126.34% (transaction costs not included)
Estimated volatility: 12.02%
Source paper:
Junqing Kang, Shen Lin, Xiong Xiong: What Drives Intraday Reversal? Illiquidity or Liquidity Oversupply?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3756630
Abstract:
We document and compare the cross-section intraday reversal patterns between Chinese and U.S. market, and further study the driving forces for each of them. Different from U.S. market, there is no significant dependence of intraday reversal on stock liquidity for Chinese market. Hence, the illiquid based explanation, which has been long documented in U.S. data, could not be automatically applied. Instead, we show there is a significant resilience pattern for accumulated returns after intraday reversal: the negative correlation between previous intraday return and future return in Chinese market is reversed afterwards. These results speak for liquidity oversupply explanation. Moreover, intraday reversal is significantly stronger for stocks with higher participation of retailed investors, providing further supporting evidence. Our results shed some new light on economic drivers for intraday return behaviors, which has important implications for empirical asset pricing tests and provides understandings for the limits of market efficiency.
#609 – Intraday Reversal in US
Period of rebalancing: Intraday
Markets traded: equities
Instruments used for trading: stocks
Complexity: Simple strategy
Backtest period: 2000-2012
Indicative performance: 22.42%
Estimated volatility: 5.41%
Source paper:
Junqing Kang, Shen Lin, Xiong Xiong: What Drives Intraday Reversal? Illiquidity or Liquidity Oversupply?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3756630
Abstract:
We document and compare the cross-section intraday reversal patterns between Chinese and U.S. market, and further study the driving forces for each of them. Different from U.S. market, there is no significant dependence of intraday reversal on stock liquidity for Chinese market. Hence, the illiquid based explanation, which has been long documented in U.S. data, could not be automatically applied. Instead, we show there is a significant resilience pattern for accumulated returns after intraday reversal: the negative correlation between previous intraday return and future return in Chinese market is reversed afterwards. These results speak for liquidity oversupply explanation. Moreover, intraday reversal is significantly stronger for stocks with higher participation of retailed investors, providing further supporting evidence. Our results shed some new light on economic drivers for intraday return behaviors, which has important implications for empirical asset pricing tests and provides understandings for the limits of market efficiency.
New research papers related to existing strategies:
#581 – Equity Index and Risk-free Asset Allocation Using Neural Networks
#496 – Using Machine Learning to Pick the Right Combination of Risky and Risk-Free Asset
Benhamou, Eric and Saltiel, David and Ungari, Sandrine and Mukhopadhyay, Abhishek: Time Your Hedge With Deep Reinforcement Learning
https://ssrn.com/abstract=3693614
Abstract:
Can an asset manager plan the optimal timing for her/his hedging strategies given market conditions? The standard approach based on Markowitz or other more or less sophisticated financial rules aims to find the best portfolio allocation thanks to forecasted expected returns and risk but fails to fully relate market conditions to hedging strategies decision. In contrast, Deep Reinforcement Learning (DRL) can tackle this challenge by creating a dynamic dependency between market information and hedging strategies allocation decisions. In this paper, we present a realistic and augmented DRL framework that: (i) uses additional contextual information to decide an action, (ii) has a one period lag between observations and actions to account for one day lag turnover of common asset managers to rebalance their hedge, (iii) is fully tested in terms of stability and robustness thanks to a repetitive train test method called anchored walk forward training, similar in spirit to k fold cross validation for time series and (iv) allows managing leverage of our hedging strategy. Our experiment for an augmented asset manager interested in sizing and timing his hedges shows that our approach achieves superior returns and lower risk.
#379 – Carry Factor in Cryptocurrencies
Franz, Friedrich-Carl and Schmeling, Maik: Crypto Carry
https://ssrn.com/abstract=3774118
Abstract:
We study carry trades in the cryptocurrency market and document that the lack of sufficient arbitrage capital in combination with highly levered speculators creates ample carry opportunities. We find that carry in bitcoin, as measured by the funding rate of BTC/USD perpetual swaps, resembles components of a fear-and-greed index and past returns. Carry in perpetual swaps positively predicts returns in the time-series of BTC/USD and in a cross-section of 51 cryptocurrencies.
#536 – Machine Learning Stock Picking
Geertsema, Paul G. and Lu, Helen: The Cross-section of Long-run Expected Stock Returns
https://ssrn.com/abstract=3774548
Abstract:
We predict cumulative stock returns over horizons from 1 month to 10 years using a tree-based machine learning approach. Cumulative stock returns are significantly predictable in the cross-section over all horizons. A hedge portfolio generates 250 bp/month at a 1 year horizon and 110 bp/month at a 10 year horizon. Individual stock returns are significantly predictable at all horizons in panel data. Cashflow and momentum related predictors are mostly important at shorter horizons while dividend yield and value related predictors are more important at longer horizons. By contrast, variables related to turnover and volatility are influential at all horizons.
#536 – Machine Learning Stock Picking
Liu, Hongyi: Deep Learning for Conditional Asset Pricing Models
https://ssrn.com/abstract=3723237
Abstract:
We propose a new pseudo-Siamese Network for Asset Pricing (SNAP) model, based on deep learning approaches, for conditional asset pricing. Our model allows for the deep alpha, deep beta and deep factor risk premia conditional on high dimensional observable information of financial characteristics and macroeconomic states, while storing the long-term dependency of the informative features through long short-term memory network. We apply this method to monthly U.S. stock returns from 1970-2019 and find that our pseudo-SNAP model outperforms the benchmark approaches in terms of out-of-sample prediction and out-of-sample Sharpe ratio. In addition, we also apply our method to calculate deep mispricing errors which we use to construct an arbitrage portfolio K-Means clustering. We find that the arbitrage portfolio has significant alphas.
#490 – Lazy Stock Prices
Sadlo, Sven-Philip: Copy-Paste Outperformance: Lazy Investors and Copied Reports
https://ssrn.com/abstract=3748216
Abstract:
Cohen, Malloy, and Nguyen (2020) show that year-over-year changes in annual and quarterly reports of US firms significantly predict future returns: ”changers” generate negative alpha and underperform ”no changers” considerably. I replicate this anomaly for an updated and investable sample of S&P 1500 constituents between 1996 and June 2020 (3,668 firms, 208,425 reports). The quintile of firms with least similar reports still generates negative 3- and 5-factor alpha of up to -5.12% per year. In contrast to Cohen et al. (2020), I find a pronounced asymmetry: ”changers” underperform but ”no changers” do not outperform. The prevailing explanation is inattention: ”lazy” investors apparently do not read reports carefully and fail to spot important changes. I provide novel evidence for this argument, as ”changers” continue to underperform in the attention-grabbing universe of S&P 500 firms. Given that ”changers” earn lower future margins and returns on invested capital, these results challenge semi-strong market efficiency for probably the most widely followed stocks in the world.
#429 – CAPE Sector Picking Strategy
#207 – Value Factor – CAPE Effect within Countries
Xie, Amy: Forecasting Long-Term Equity Returns: A Comparison of Popular Methodologies
https://ssrn.com/abstract=3774998
Abstract:
Many investors need to make long-term asset class forecasts for planning and portfolio construction purposes. We examine the empirical performance of two different approaches to forecasting future ten-year equity returns: a regression methodology using CAPE and a more traditional “building block” approach. The regression approach produces estimates that are poor predictors of subsequent actual returns. The “building block” approach (BBA) outperforms the regression methodology (in terms of root mean squared error) with the repricing component helping to capture periods of poor equity returns. A high CAPE value is not necessarily cause for alarm and changes in asset allocation. If an investor plans to use a methodology that over time will prove more accurate, then the historical record is more supportive of the BBA approach, with or without a repricing component based on current P/E.
#513 – Predicting Intraday Returns with Machine Learning methods
Roh, Hyunwoo: Predicting High-frequency Stock Price Using Machine Learning Technique
https://ssrn.com/abstract=3744765
Abstract:
This paper addresses the problem of predicting the stock price using the high frequency data based on a machine learning approach. We study two things in this paper (1) comparison of the prediction performance among selected function classes with given look-back parameter in terms of the proposed evaluation measures in the process of finding the best in-sample empirical loss minimizer (2) the comparison of those results by changing the sampled frequency of financial time series data after obtaining an introduced set of high frequency data features extracted from the Trades and Quotes (TAQ) data. For the analysis of TAQ data, feature engineering involves the computation of 56 number of related features including market microstructure, statistical, and technical indicator features. Re-estimation was done to improve the prediction accuracy for data models to obtain the predicted value every moving window. On the other hand, algorithmic models are used without re-estimation for the practical matter in that the time spent to train the model is often larger than the sampled frequency of the data. Moreover, the look-back parameter is introduced to cut off the irrelevant long past historical data. Among the selected function class in the experiment, the results show that the PCA regression performs the best in terms of the mean directional accuracy and simple back testing for both NASDAQ100 index and TAQ data with given sampled frequencies (i.e., 3min, 5min, etc). Compared to previous studies using NASDAQ100, the results demonstrate that re-estimation and properly chosen look-back parameter improve the prediction performance in terms of proposed evaluation measures. When it comes to maximum draw down, which is a measure critical for risk management, DA-RNN rendered the smallest value and, thereby, was the best performing model for TAQ data for all time frequencies. We also provide DM statistics whose null hypothesis is that the accuracies of prediction values of any two given models will not be different. In case of TAQ data for all sampled frequencies, there is evidence that we cannot reject the null hypothesis when comparing between PCA regression and DA-RNN model. Extensive experiments provide insights into properly evaluating the prediction performance of best in-sample empirical loss minimizer using the high frequency time series data.
#541 – News and Non-news Returns in Stocks
Dang, Van and Liu, Leo: Easy to Parse, Easy to Trade
https://ssrn.com/abstract=3792720
Abstract:
The delivery of information is as important as the content itself. We present evidence that an announcement’s writing literacy is as critical as its information content. We show that the text’s readability in firm-specific news affects trading decision, particularly those of algorithmic traders. We find that algorithmic participants trade more aggressively when the news items are more readable. This change in their trade decisions leads to several market impacts. First, we suggest that readability affects algorithmic traders’ liquidity provision. Algorithmic traders can capture more rents when the news is more readable while other participants enjoy lower adverse selection costs. The market makers’ profit is on average less than the reduced selection costs. Therefore, there is an improvement in overall market welfare in terms of the transaction cost. Finally, price efficiency is improved when the market can trade on highly readable news.
#513 – Insiders Trading Effect in Stocks
Patel, Vinay and Putnins, Talis J.: How Much Insider Trading Happens in Stock Markets?
https://ssrn.com/abstract=3764192
Abstract:
We estimate that the actual prevalence of illegal insider trading is at least four times greater than the number of prosecutions. Using novel structural estimation methods that explicitly account for the incomplete and non-random detection and hand-collected data of all US prosecuted insider trading cases, we estimate that insider trading occurs in one in five mergers and acquisition events and in one in 20 earnings announcements. Key drivers of the decision to engage in illegal insider trading include stock liquidity, the value of the inside information, and the number of people in possession of the information. Detection and prosecution are more likely when there are abnormal trading patterns and more regulatory resourcing.
#294 – Seasonality Within Trend-Following Strategy in Commodities
Ewald, Christian-Oliver and Haugom, Erik and Lien, Gudbrand and Størdal, Ståle and Wu, Yuexiang: Trading Time Seasonality in Commodity Futures: An Opportunity for Arbitrage in the Natural Gas and Crude Oil Markets?
https://ssrn.com/abstract=3792028
Abstract:
For fixed maturity, under the no-arbitrage assumption, futures prices should follow a martingale with respect to the trading time, at least under the pricing measure. Therefore, a prominent display of trading time seasonality under the physical measure raises warning signs and can only occur by means of strong seasonality in the pricing kernel. We show that for natural gas and crude oil, trading time seasonality is present to an extent where it may violate the no-arbitrage assumption. We provide three layers of evidence. The first layer is descriptive only, the second involves the Kruskal–Wicksell test for establishing trading time seasonality, and the third is in the form of a trading strategy, which exploits trading date seasonality. This strategy can produce statistically significant positive alphas in the CAPM context, thereby indicating the possibility of an arbitrage.
And two interesting free blog posts have been published during last 2 weeks:
An Investigation of R&D Risk Premium Strategies
The R&D investments represent a company’s unique expenditure, which is responsible for creating an information asymmetry about the firm’s growth potential and future prospects. In a case when market value reflects only the firm’s financial statements without taking the long-term benefits of R&D investments into consideration, the company’s stocks may be underpriced. On the other hand, the firm’s stock prices may also face overpricing. This might happen in a case when the investors judge the possible future outcomes of current R&D investment based on the past firm’s R&D success, which is not a guarantee by any means.
So, is there a premium among firms with intensive expenditures on R&D or not? If so, does R&D expenditures represent a robust risk factor, or are there any other hidden economic forces that could explain the R&D premium? This article has tried to answer these questions by revisiting and expanding the three previously conducted research papers on R&D premium.
Retail Investment Boom, Robinhood, Passive Investing and Market Inelasticity
This week’s blog is unique compared to our previous posts. We have identified two papers that are connected, each with interesting findings and implications. One of today’s leading topics is the Robinhood trading platform, but not from the point of view of recent short squeezes and speculations. The Robinhood can be an interesting insight into retail investing and implications for the market. Research suggests that despite the very low share of retail investors, their power is significantly high. This seems to be caused by the inelastic market, which passive investing contributes to. Therefore, inelasticity is another crucial point.
Plus, the following six trading strategies have been backtested in QuantConnect in the previous two weeks:
#258 – FX Value v2 – Real Exchange Rate Changes
#259 – FX Value v3 – Real Exchange Rate Levels
#498 – Value in Anomalies
#534 – Time Series Factor Momentum
#578 – Combining Smart Factors Momentum and Market Portfolio
#603 – Slope Carry
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
