A new study analyses correlations between six cryptocurrencies and the relationship between cryptos as well as compared to fiat pairs, revealing some interesting results.
Covering an extensive period from the first of September 2015 to as late as the 29 December 2017, the study generally finds week to no correlation between cryptos and fiat pairs, but some appear to be correlating more than others.
As can be seen above, there is strong correlation between fiat pairs, especially between the euro and swiss franc, with the Japanese Yen and the pound the only ones to show no correlation.
Generally, cryptos show no correlation to fiat pairs, but there does appear to be an inverse correlation relationship between bitcoin and the pound, the Canadian dollar, which usually acts like the USD, as well as bitcoin and the Yuan.
That is, when the dollar gets stronger bitcoin tends to fall somewhat, and likewise when it gets weaker it tends to rise, but the relationship isn’t that strong at just 0.1 out of 1 for complete correlation and 0 for none.
Ethereum seems to have the strongest relationship with the pound in fiat pairs, but that relationship is weaker than bitcoin, with the other crypto/fiat correlations too weak to be significant.
Between cryptos there is a fairly strong correlation, especially between bitcoin and litecon as well as ripple and stellar. But a moving-average cross-correlation analysis reveals even more interesting results:
Here, we quote the study authors at some length as they eloquently explain the relationships between cryptos and fiats:
“We found a high correlation between BTC and LTC (0.48). Now, we can see that when both BTC and LTC record extreme negative returns (0.05|0.05 quantiles), the coherency is 0.46.
When both record extreme positive returns (0.95|0.95 quantiles), the coherency is 0.27. In median dependence (0.50|0.50 quantiles), the coherency is 0.7. Thus, these two cryptocurrencies are connected to some extent in the short- term period, irrespective of the current market situation.
What is interesting is that XEM is on average not associated with any other currency in our sample. However, when we look at the lower quantiles, we find that it is negatively associated with two forex currencies (JPY, CHF) and positively associated with LTC.
At the higher quantiles, its extreme positive returns are positively connected to extreme positive returns on EUR, GBP, CAD, and XLM and negatively connected with BTC.
Even ETH is significantly positively associated with BTC at the extreme lower quantile and median dependence, even though there are substantial differences between these two cryptocurrencies.”
Some cryptos, therefore, do appear to have a slight hedging quality, especially against the pound and the dollar, but generally the study reveals they are useful for diversification between crypto/fiat and between cryptos themselves.
Interestingly the study says XRP is the one that exhibits safe heaven properties, which does seem to have the lowest correlation between cryptos.
As has been seen from observation, the study suggests cryptos, and LTC in particular, tend to respond more towards bitcoin’s price falls, and half as less towards bitcoin’s price rises. But there are significant differences, leading the authors to conclude after further analysis:
“It is beneficial to diversify among forex currencies and cryptocurrencies because in times of distress, extremely low returns are negatively associated.”
A general body of research now seems to be agreeing that cryptos are useful for diversification, with a further study recently finding no correlation between bitcoin and a number of assets such as stocks, gold or oil.
That suggests even a small amount of cryptocurrencies in traditional portfolios aids in increasing returns due to spreading and diversifying risks as well as gains.
Comment