Purpose: The present study sets out to examine the empirical literature on the behavioural aspects of cryptocurrencies, showing the findings of related studies and discussing the various results. A systematic literature review of cryptocurrencies in behavioural finance seems to be timely and particularly important in terms of providing a guide for future research. Key topics include an extent review on the issue of herding behaviour amongst cryptocurrencies, momentum effects and overreaction, contagion effect, sentiment and uncertainty, along with studies related to investment decision-making, optimism bias, disposition, lottery and size effects.Design/methodology/approach: Systematic literature review.Findings: A systematic literature review of cryptocurrencies in behavioural finance seems to be timely and particularly important in terms of providing a guide for future research. Key topics include an extent review on the issue of herding behaviour amongst cryptocurrencies, momentum effects and overreaction, contagion effect, sentiment (investor's, market's) and uncertainty, along with studies related to investment decision-making, optimism bias, disposition, lottery and size effect.Originality/value: The authors' survey paper complements recent papers in the area by offering a systematic account on the influence of behavioural factors on cryptocurrencies. Further, this study's purpose is not just to index the relevant literature, but rather to showcase and pinpoint several research areas that have emerged in the field of behavioural cryptocurrency research. For all these reasons, a systematic literature review of cryptocurrencies in behavioural finance seems to be timely and particularly important.

In this paper, we present two neural-network-based techniques: an adaptive evolutionary multilayer perceptron (aDEMLP) and an adaptive evolutionary wavelet neural network (aDEWNN). The two models are applied to the task of forecasting and trading the SPDR Dow Jones Industrial Average (DIA), the iShares NYSE Composite Index Fund (NYC) and the SPDR S&P 500 (SPY) exchange-traded funds (ETFs). We benchmark their performance against two traditional MLP and WNN architectures, a smooth transition autoregressive model (STAR), a moving average convergence/divergence model (MACD) and a random walk model. We show that the proposed architectures present superior forecasting and trading performance compared to the benchmarks and are free from the limitations of the traditional neural networks such as the data-snooping bias and the time-consuming and biased processes involved in optimizing their parameters. Copyright © 2015 John Wiley & Sons, Ltd.