Cryptocurrencies

Recently, many investors have become interested in investing in cryptocurrency market. Investing in an asset carries a lot of risk and may bankrupt the investor. The main way to control this risk is portfolio diversification. In this paper, we will investigate the effect of portfolio diversification by adding cryptocurrencies to the portfolio. We evaluate the performance of seven risk-based portfolio optimization strategies. these strategies are the minimum variance, inverse volatility, L2-norm constrained minimum variance, L2-norm constrained maximum decorrelation, risk parity portfolio and maximum diversification. Our portfolios consist of three markets stocks including, Tehran Stock Exchange, Commodities and Cryptocurrencies. Also, due to the fact that the cryptocurrency market has gained a significant attraction among investors, we will examine the positive and negative effects of adding the five selected currencies, simultaneously and separately to the base portfolio, which is Tehran Stock Exchange-Commodities portfolio. We investigate that whether adding cryptocurrencies to a stock portfolio can be considered as a tool to improve a risk-based portfolio. After analyzing portfolios, the best portfolio in each strategy and the best strategy in each portfolio are introduced from the aspects of risk, return and Sharpe ratio, and finally we have concluded that entering the cryptocurrency market in most of the strategies lead to an overall increase in the return, while the approach is to minimize the risk of the portfolio. So, it can be concluded that if the main goal is to build a more diversified portfolio, better outcome can be obtained for the investor considering the return gained.

This paper intends to model the volatilities of returns of 20 different cryptocurrencies using daily data from 08/03/2018 to 09/20/2022. The multivariate factor stochastic volatility model (MFSV) within the framework of the nonlinear space-state approach is used. In this method, the cryptocurrency return volatility is decomposed into volatility rooted in latent factors and idiosyncratic volatility, and the time-varying pairwise correlation and dynamic covariance matrix are estimated in four sub-periods. The MFSV model’s results revealed that each sub-period contains a distinct number of latent factors, 2, 5, 4 and 2, which generally have a favorable impact on all cryptocurrency volatilities. The time-varying positive correlations between the return volatility of all cryptocurrencies are confirmed. Indeed, the strongest pairwise correlations belong to Ethereum, Litcoin, EOS, and VET in each sub-period, respectively. The DOGE, DOGE, Filecoin, and XRP, on the other hand, showedthe weakest correlations . As the pairwise correlations of cryptocurrency volatilities get strenger, especially during descending periods, it seems that the benefits of diversifying a crypto portfolio are getting less and less over time.