*6.2. Future Works*

Understanding how cryptocurrencies are correlated with each other sheds light on portfolio optimization. Based on the outcome of this study, we can take one step further by constructing and comparing the portfolio optimizations at different phases of the market, i.e., during bear and bull market periods. Therefore, the unique characteristics of an optimized portfolio at different market phases can, in theory, be learned and analysed. Secondly, we have noted that different network structures can be observed for a number of exchanges. Thereby, a comparison between them can be made. Another future plan which is worth taking into consideration is to observe the correlation using different techniques. For instance, we are aiming to use mutual information, which is successfully applied in [25], to estimate the correlation between two cryptocurrencies. This method can overcome obstacles from popular linear and non-linear methods since it can measure the non-linear correlation while allowing the existence of non-monotonic relationships. Lastly, we have noticed that the network structure of low-frequency data behaves differently to that of high-frequency data. We remark that we can expect to learn the long-term characteristics of cryptocurrencies based on this structure which could be potentially beneficial for investors who choose to make a long-term investment decision.
