We retrieve news stories and earnings announcements of the S&P 100 constituents from two\ud professional news providers, along with tenmacroeconomic indicators.We also gather data fromGoogle\ud Trends about these firms’ assets as an index of retail investors’ attention. Thus, we create an extensive\ud and innovative database that contains precise information with which to analyze the link between news\ud and asset price dynamics. We detect the sentiment of news stories using a dictionary of sentiment-related\ud words and negations and propose a set of more than five thousand information-based variables that\ud provide natural proxies for the information used by heterogeneousmarket players. We first shed light on\ud the impact of information measures on daily realized volatility and select them by penalized regression.\ud Then,we performa forecasting exercise and showthat themodel augmentedwith news-related variables\ud provides superior forecasts

Bitcoin is foremost amongst the emerging asset class known as cryptoassets. Two noteworthy characteristics of the returns of nonstablecoin cryptoassets are their high volatility, which brings with it a high level of risk, and their high intraclass correlation, which limits the benefits that can be had by diversifying across multiple cryptoassets. Yet cryptoassets exhibit no correlation with gold, a highly-liquid yet scarce asset which has proved to function as a safe haven during crises affecting traditional financial systems. As exemplified by Shannon's Demon, a lack of correlation between assets opens the door to principled risk control through so-called volatility harvesting involving periodic rebalancing. In this paper we propose an index which combines a basket of five cryptoassets with an investment in gold in a way that aims to improve the risk profile of the resulting portfolio while preserving its independence from mainstream financial asset classes such as stocks, bonds and fiat currencies. We generalise the theory of Equal Risk Contribution to allow for weighting according to a desired level of contribution to volatility. We find a crypto-gold weighting based on Weighted Risk Contribution to be historically more effective in terms of Sharpe Ratio than several alternative asset allocation strategies including Shannon's Demon. Within the crypto-basket, whose constituents are selected and rebalanced monthly, we find an Equal Weighting scheme to be more effective in terms of the same metric than a market capitalisation weighting.

No abstract

No abstract