Crypto Culture: Should You Diversify Your Crypto Portfolio?

Already available on Bloomberg, Reuters and TradingView, the CIX100 index of the 100 most active cryptocurrencies has finally landed on the Nasdaq. So there’s no time like the present to talk about the relevance of diversity in your crypto portfolio. 

What is the CIX 100 index?

The CIX 100 is an AI-based cryptocurrency index — see cryptoindex.io or cryptoindex.com. Designed to represent the “tendency” of the market by grouping together the top 100 cryptocurrencies, its permanent update calculation is based on the proprietary algorithm, Zorax.

Using artificial intelligence processes (including natural language processing and machine learning), Zorax processes information collected from 9 exchange platforms and on more than 1,800 cryptocurrencies in real-time.

Having raised over $11 million in 2019, Cryptoindex’s Russian technical director, Sergey Sheshev, has revealed very little about the particular routines and processes on which the index is based, leaving many to speculate on what’s going on behind the scenes.

Taking a look at cix100.com, we can, however, see that the index gives a heavy weighting to Bitcoin (more than 70% of the value), and claims a performance of 1105% — though this does include the huge bull run of 2017, so ought to be taken with a pinch of salt.

The need for a cryptocurrency index is clear: it perpetuates the asset class by allowing it to be compared with others (Dow Jones Industrial Average-DJINDICES:^DJI, Nasdaq Composite-NASDAQINDEX:^IXIC, S&P 500-SNPINDEX:^GSPC, etc.).

This is essential to be able to study the correlations between different asset classes when looking to optimize your portfolio. Although other alternatives are available (such as the NCI indexes offered by netcurrencyindex.com), the CIX100 benefits from the privileged position of being first in its category, and can pride itself on the approbation of its most prestigious peers.

Index CIS100 whose composition is available on cix100.com:

"Modern Portfolio Theory"

There are many different approaches to the construction of an asset portfolio. For instance, we can establish its list of assets from the fundamental analysis of each proposed asset, refer instead to technical analysis criteria (market asset prices), or base its calculations on sophisticated proprietary ratios. Or, simply, buy an index.

Whatever the approach, the basic idea remains the same: to diversify your investment and limit your risk, because, as everyone knows, "you never put all your eggs in one basket”. 

This model is widely credited to American economist Harry Markowitz who, in 1952, published his article “Portfolio Selection” in the Journal of Finance, which outlined his “modern portfolio theory”. A mean-variance analysis framework which received the Nobel Prize in 1990, Markowitz’s approach allows you to precisely calculate the maximal possible performance for a portfolio asset via: the correlation rate between assets; their variance; and the risk to which the investor is willing to commit (expressed by the standard deviation).

The result is the “Markowitz frontier” which expresses all portfolio combinations (i.e. asset allocations), delivering the best performance for given overall portfolio risk. 

Although taught for decades in Business Schools, modern portfolio theory has since been staunchly criticized because it is considered, although mathematically coherent, to be disconnected from the reality of financial markets (see Nassim Nicholas Taleb's The Black Swan Theory, and other crash risks that are not taken into account by the MPT), however a mix of low-correlation assets remain the key to all investment strategies.

Calculation scheme of the optimal portfolio according to the "Modern Portfolio Theory", proposed by Harry Markowitz in 1952 - (F. Bonelli - TheCrypto MBA)

Crypto Correlation / Decorrelation

A global portfolio is generally constituted of several weakly-correlated asset classes, with the hope that the temporary performance peaks of some will compensate for the temporary underperformance troughs of the others. But to be able to create such a portfolio, we first need to verify that there is a low correlation between assets of the same class within the portfolio.

To clarify, we define correlation as a measure of the statistical tendency for variables to behave in the same way at the same time. It is generally expressed as a value between -1 and +1 (-1 = perfect opposition, +1 = perfect correlation, and 0 = no correlation between the variables).

Simply, we want to work with assets that are either in opposition or are weakly correlated, in order to offset losses with gains elsewhere. However, things are a bit more complicated than that — after all we want to structure a profitable portfolio.

To do this with cryptocurrency, it is also necessary to structure the weights of each selected crypto, and actively manage our selection because the correlation values vary over time.

Studies now exist on the correlation between cryptocurrencies which show that “altcoins” remain strongly correlated to Bitcoin during a “bull run” (strong rise in prices). The situation is quite different when the market oscillates or falls. In the same vein, a recent study published by Binance seems to show that a stronger correlation has been established between Ethereum and altcoins than between Bitcoin and the altcoins. 

The results show that a portfolio judiciously constituted of coins selected from the Top 100, and frequently rebalanced (weekly) will be widely more performant than simply “hodling a Bitcoin bag” (a package of Bitcoins in passive management without trading). And this is precisely the value proposition of cryptocurrency investment funds.

But before you jump in, be aware of one thing. The study of correlations and the more “efficient” calculus of the portfolio requires access to large datasets and a number of complex calculations that are generally beyond the reach of an individual.

That’s not to mention the management of trading operations that allow the portfolio to be constantly adjusted (rebalancing, trading cryptocurrencies, etc.).

And it’s for this reason that an index like the CIX100 has real value — it does all this work for you.

Market Making

A word of warning. An index like the CIX100, which has been issued by a private company and which makes recommendations for the purchase and sale of assets, may be misused.

If investors and institutions are allowed to affect index management decisions, it will quickly become a formidable market-making tool, able to use the volume of transactions it generates through its users to "make the market" on its own.

For example, if the index indicates that it will take a cryptocurrency out of the market and makes a downward recommendation, a large numbers of speculators are going to follow this recommendation and sell, which will cause the rate of the cryptocurrency to drop sharply and make the position of the index “self-predictive”: as long as some smart guys anticipated the phenomenon, they only need to “short” their positions on the coin (i.e. focus on lowering the price through short selling) to collect a comfortable benefice on the basis of a price movement that they have triggered...

This sort of market manipulation is endemic to the cryptosphere, and only adds to the sulphurous reputation of cryptocurrencies because it is still very poorly regulated (and punished).

For instance, many will turn to Telegram groups, which are very popular in the crypto-world, for clues to market trends, but many of these groups are attempting to make a market out of budding investors too. You think you have privileged access to market information by subscribing to a paid group, while you will, in fact, be used as a servile executor as part of the market-making strategy which allows the group's administrators to enrich themselves on your back.

Investing in cryptocurrencies is no leisurely cruise.