Noise traders,firm-specific uncertainty and technical trading effectiveness

This is the first study to document evidence of technical trading effectiveness at firm level in the Chinese A-share market by investigating the relationship between excess profits of technical trading rules and firm-specific characteristics. Our results reveal that firms with higher excess profits from technical trading have more noise traders and higher institutional ownership and that those firms tend to be growth firms with lower liquidity and higher firm-specific uncertainty. Further analysis shows that the profitability of technical trading rules is unsustainable and the excess profits of the highest technical trading profit quintile portfolio disappear in the following year.     

Size and information revelation in securities trading

We examine differences in information content between order submission sizes and trade sizes by U.S. equity traders. Increasing (decreasing) order submission (trade) size is reflective of information. The result suggests that better-informed traders want to trade in a large size, but that they engage in stealth trading practices or break larger orders into smaller sizes in order to conceal information. While prior studies tend to narrowly focus on trade executions at the market-centre level, our findings indicate that order submission size varies significantly from trade size and that both sizes are informative about future prices, albeit in an inverse manner.     

High frequency momentum trading with cryptocurrencies

Over the past few years, cryptocurrencies have increasingly been discussed as alternatives to traditional fiat currencies. These digital currencies have garnered significant interest from investment banks and portfolio managers as a potential option to diversify the financial risk from investing in other assets. This interest has also extended to the general public who have seen cryptocurrencies as a way of making a quick profit. This paper provides a first insight into the applicability of high frequency momentum trading strategies for cryptocurrencies. We implemented two variations of a signal-based momentum trading strategy: (i) a time series method; (ii) a cross sectional method. These strategies were tested on a selection of seven of the largest cryptocurrencies ranked by market capitalization. The results show that there exists potential for the momentum strategy to be used successfully for cryptocurrency trading in a high frequency setting. A comparison with a passive portfolio strategy is proposed, which shows abnormal returns when compared with the momentum strategies. Furthermore, the robustness of our results are checked through the application of the momentum strategies other sample periods. We also compare the performances of the signal-based momentum strategies with returns-based versions of the strategies. It is shown that the signal-based strategy outperforms the returns-based strategy. However, there appears to be no single parameterization of the signal-based strategies that can generate the greatest cumulative return over all sample periods.     

Data vs. information: Using clustering techniques to enhance stock returns forecasting

This paper explores the use of clustering models of stocks to improve both (a) the prediction of stock prices and (b) the returns of trading algorithms.We cluster stocks using k-means and several alternative distance metrics, using as features quarterly financial ratios, prices and daily returns. Then, for each cluster, we train ARIMA and LSTM forecasting models to predict the daily price of each stock in the cluster. Finally, we employ the clustering-empowered forecasting models to analyze the returns of different trading algorithms.We obtain three key results: (i) LSTM models outperform ARIMA and benchmark models, obtaining positive investment returns in several scenarios; (ii) forecasting is improved by using the additional information provided by the clustering methods, therefore selecting relevant data is an important preprocessing task in the forecasting process; (iii) using information from the whole sample of stocks deteriorates the forecasting ability of LSTM models.These results have been validated using data of 240 companies of the Russell 3000 index spanning 2017 to 2022, training and testing with different subperiods.     

Combining wavelet decomposition with machine learning to forecast gold returns

This paper combines the discrete wavelet transform with support vector regression for forecasting gold-price dynamics. The advantages of this approach are investigated using a relatively small set of economic and financial predictors. I measure model performance by differentiating between a statistically-motivated out-of-sample forecasting exercise and an economically-motivated trading strategy. Disentangling the predictors with respect to their time and frequency domains leads to improved forecasting performance. The results are robust compared to alternative forecasting approaches. My findings on the relative importances of such wavelet decompositions suggest that the influences of short-term and long-term trends are not stable over the full evaluation period.     

Classification of intraday S&P500 returns with a Random Forest

Stock markets can be interpreted to a certain extent as prediction markets, since they can incorporate and represent the different opinions of investors who disagree on the implications of the available information on past and expected events and trade on their beliefs in order to achieve profits. Many forecast models have been developed for predicting the future state of stock markets, with the aim of using this knowledge in a trading strategy. This paper interprets the classification of the S&P500 open-to-close returns as a four-class problem. We compare four trading strategies based on a random forest classifier to a buy-and-hold strategy. The results show that predicting the classes with higher absolute returns, ‘strong positive’ and ‘strong negative’, contributed the most to the trading strategies on average. This finding can help shed light on the way in which using additional event outcomes for the classification beyond a simple upward or downward movement can potentially improve a trading strategy.     

Performance and learning in an ambiguous environment: A study of cryptocurrency traders

We investigate the performance and learning ability of traders in an environment governed by ambiguity, such as the cryptocurrency market. Using a profit decomposition methodology, we find significant cross-sectional and temporal heterogeneity in performance. Traders do not learn to progressively increase the magnitude of returns; however, they are able to improve on their ability to realise profits as a mechanism of adaptation to survive through ambiguity. This adaptation increases as traders progress through their career. Moreover, we find evidence in support of the gambler’s fallacy. We argue that learning in ambiguous environments has limitations, allowing traders primarily to survive.     

Smoking hot portfolios? Trading behavior,investment biases,and self-control failure

Self-control is a personality trait that explains undersaving and nonparticipation decisions. We show that self-control failure also affects trading behavior among individuals on capital markets. We use smoking as the most socially accepted example of self-control failure among 13,644 German brokerage clients and compare the trading behavior of 3,553 smokers and 10,091 nonsmokers. Smokers are associated with a higher portfolio turnover unexplained by financial sophistication or wealth effects. Self-control failure also exacerbates overconfidence, social contagion, sensation seeking, and attention grabbing. Overall, self-control failure is costly because it increases the gap between gross and net returns of smokers relative to nonsmokers.     

Data revisions to German national accounts: Are initial releases good nowcasts?

Data revisions to national accounts pose a serious challenge to policy decision making. Well-behaved revisions should be unbiased, small, and unpredictable. This article shows that revisions to German national accounts are biased, large, and predictable. Moreover, with use of filtering techniques designed to process data subject to revisions, the real-time forecasting performance of initial releases can be increased by up to 23%. For total real GDP growth, however, the initial release is an optimal forecast. Yet, given the results for disaggregated variables, the averaging out of biases and inefficiencies at the aggregate GDP level appears to be good luck rather than good forecasting.     

When do retail investors pay attention to their trading platforms?

Buying and selling securities through online trading platforms has become increasingly popular among U.S. households in recent years. This study tracks U.S. households' attention to their online trading platforms using daily data for 2004 to August 2017. The analysis covers the 10 most popular online trading platforms among U.S. investors. The findings indicate that market shocks, captured by several proxies, as well as macroeconomic announcements attract investors' attention to trading platforms. We also document that the ostrich effect weakens when considering greater changes in the VIX. Our findings do not support the avoidance of information theory, but do support the theoretical argument that risk-averse agents engage in more information gathering when uncertainty prevails in hopes of reducing their risks.