Z-score vs minimum variance preselection methods for constructing small portfolios

The rise of robo-advisor wealth management services, which constitute a key element of fintech revolution, unveils the question whether they can dominate human-based advice, namely how to address the client’s behavioral biases in an automated way. One approach to it would be the application of machine learning tools during client profiling. However, trained neural network is often considered as a black box, which may raise concerns from the customers and regulators in terms of model validity, transparency, and related risks. In order to address these issues and shed more light on how neurons work, especially to figure out how they perform computation at intermediate layers, this paper visualizes and estimates the neurons’ sensitivity to different input parameters. Before it, the comprehensive review of the most popular optimization algorithms is presented and based on them respective data set is generated to train convolutional neural network. It was found that selected hidden units to some extent are not only specializing in the reaction to such features as, for example, risk, return or risk-aversion level but also they are learning more complex concepts like Sharpe ratio. These findings should help to understand robo-advisor mechanics deeper, which finally will provide more room to improve and significantly innovate the automated wealth management process and make it more transparent.

Retracted on August 17, 2020 by the Journal’s owner and Publisher. Type of retraction – plagiarism.
There wasn’t a request for this retraction, but the reason for investigation of plagiarism fact was the Russian Academy of Sciences Committee’s report “Predatory Journals at Scopus and WoS: Translation Plagiarism from Russian Sources”: https://kpfran.ru/wp-content/uploads/plagiarism-by-translation-2.pdf” dated August 12, 2020. The publishing house has familiarized itself with the report. The article by Alexey Mikhaylov, Natalia Sokolinskaya and Evgeniy Lopatin (2019). Asset allocation in equity, fixed-income and cryptocurrency on the base of individual risk sentiment. Investment Management and Financial Innovations, 16(2), 171-181. doi:10.21511/imfi.16(2).2019.15 was mentioned in this report. It is noted that translation plagiarism was detected in this article - http://wiki.dissernet.org/wsave/IMFI_2019_2_1publ.html.
Due to this the publishing house carried out an investigation on possible cases of plagiarism of all articles of these authors (Alexey Mikhaylov, Natalia Sokolinskaya and Evgeniy Lopatin) published in “Business Perspectives” journals.
When the manuscript "Asset allocation in equity, fixed-income and cryptocurrency on the base of individual risk sentiment" was submitted to the Journal for consideration, the authors signed the Cover letter and attested to the fact that their manuscript is an original research and has not been published before. Then, the manuscript was accepted for consideration by the Managing Editor and was tested for plagiarism using the iThenticate and Unicheck programs. Plagiarism was not detected. On August 12, 2020 the Russian Academy of Sciences Committee’s presented the report. Editorial staff decided to re-test all articles of mentioned authors for plagiarism using the iThenticate and Unicheck programs – the programs didn’t show the plagiarism, then the articles were tested for translation plagiarism by the experts of “Business Perspectives” and plagiarism was detected (plagiarism and paraphrases from Russian-language sources).

According to the results of the investigation, the Publisher and owner of the journal decided to retract this article because of plagiarism on August 17, 2020.
The authors were notified of such a decision.

The asset management sector is constantly looking for a reliable investment strategy, which is able to keep its promises. One of the most used approaches is the target volatility strategy that combines a risky asset with a risk-free trying to maintain the portfolio volatility constant over time. Several analyses highlight that such target is fulfilled on average, but in periods of crisis, the portfolio still suffers market’s turmoils. In this paper, the authors introduce an innovative target volatility strategy: the discontinuous target volatility. Such approach turns out to be more conservative in high volatility periods. Moreover, the authors compare the adoption of the VIX Index as a risk measure instead of the classical standard deviation and show whether the former is better than the latter. In the last section, the authors also extend the analysis to remove the risk-free assumption and to include the correlation structure between two risky assets. Empirical results on a wide time span show the capability of the new proposed strategy to enhance the portfolio performance in terms of standard measures and according to stochastic dominance theory.