In this paper, we first obtain a time-varying measure of volatility connectedness involving fifteen major cryptocurrencies based on a dynamic conditional correlation-generalized autoregressive conditional heteroscedasticity (DCC-GARCH) model, and then analyze the role of investor sentiment in explaining the movement of the connectedness metric within a quantile-on-quantile framework. Our findings show that lower quantiles of investor happiness, built on Twitter feed data as a proxy for investor sentiment, is positively associated with the entire conditional distribution of connectedness, but the opposite is observed at higher values of investor happiness. In addition, when we look at the effect of sentiment on the common market volatility, we are able to deduce that as investors become exceedingly unhappy, overall market volatility increases and this is associated with high market connectedness. The heightened volatility possibly due to higher trading, seems to suggest that cryptocurrencies are used for hedging when investor sentiment is weak, with evidence in favor of this behavior being relatively stronger than the possible speculative motive associated with happy investors, as low total connectedness is coupled with high common volatility. Our results tend to suggest that, relatively more diversification opportunities are available when investors are happy rather than when sentiment is weak.

在本文中，我们首先基于动态条件相关-广义自回归条件异方差（DCC-GARCH）模型获得涉及15种主要加密货币的波动率连通性的时变度量，然后分析投资者情绪在解释市场波动中的作用分位数上框架内的连通性度量。我们的研究结果表明，以Twitter提要数据为代表的投资者情绪为基础的较低的投资者幸福感与整个联系度的条件分布呈正相关，但是在较高的投资者幸福感上却相反。此外，当我们观察情绪对共同市场波动的影响时，我们可以推断出，随着投资者变得极为不满，整体市场波动性增加，这与高度的市场连通性有关。可能由于较高的交易而增加的波动性似乎表明，当投资者情绪疲软时，可使用加密货币进行套期保值，有证据表明，由于总联系度低，这种行为比与快乐投资者相关的可能的投机动机相对更强。具有较高的常见波动性。我们的结果倾向于表明，当投资者感到高兴而不是情绪低落时，可获得相对更多的多元化机会。有证据表明，这种行为比与快乐投资者有关的可能的投机动机相对要强，因为总联系度低，共同波动性高。我们的结果倾向于表明，当投资者感到高兴而不是情绪低落时，可获得相对更多的多元化机会。有证据表明，这种行为比与快乐投资者有关的可能的投机动机相对要强，因为总联系度低，共同波动性高。我们的结果倾向于表明，当投资者感到高兴而不是情绪低落时，可获得相对更多的多元化机会。