Are you a visual learner?
Does your brain turn to mush when you see gigantic data tables, or huge blocks of text?
I am a visual learner, and this is one of the reasons why I love correlation matrices. We can take a simple concept and display it in an easy to digest, visually stimulating graphic.
Have you ever noticed that when the price of Bitcoin goes up or down, that a large majority of the market seems to follow?
Or, when the price of one cryptocurrency rises, another falls almost in lockstep?
This is highly apparent in the cryptocurrency world, as many of the currencies on the market are completely coupled with the movement of Bitcoin. Certain coins will move incredibly close to Bitcoin, or act as a hedge against falling prices.
In the financial world, correlation is a statistical measure of how two securities move in relation to each other.
Currency correlation, then, tells us whether two currency pairs move in the same, opposite, or comnpletely random direction, over a designated period of time.
Unless you plan on trading just one pair at a time, it’s crucial that you understand how different currency pairs move in relation to each other.
Correlation is computed into what is known as the correlation coefficient, which ranges between -1 and +1.
Perfect positive correlation (a correlation coefficient of +1) implies that the two currency pairs will move in the same direction 100% of the time.
Perfect negative correlation (a correlation coefficient of -1) means that the two currency pairs will move in the opposite direction 100% of the time.
If the correlation is 0, the movements between two currency pairs are said to have ZERO or NO correlation, they are completely independent and random from each other. This basically means that we have no real idea how one cryptocurrency will react to the movement of another.
Take a look at the following table:
I have plotted the 1-Day correlation coefficients of 12 different cryptocurrencies. Immediately we can notice a few things:
Note: This is merely for data shown on 11/13, and can change rapidly at any given time. You can also see different correlations by time period, where two currencies can be positively correlated in one-minute intervals, but negatively correlated in one-day intervals
Now that we have taken the time to learn about correlations, let's explore how to use them to our advantage.
Let's say, hypothetically that BTC/USD and EOS/USD had a perfect -1.0 correlation (ie, they move in opposite directions 100% of the time).
If you were to invest in both, this would be the same as having virtually no position. When BTC/USD rallies, EOS/USD would undergo a selloff.
On the other hand, going long in BTC/USD and shorting EOS/USD would be effectively doubling your position, since the correlations are so strong.
Bitcoin Cash (BCH) and Dash (DSH) in the image above have a correlation of 0.59.
Instead of a user buying 100% worth of BCH, they could diversify their risk somewhat while still maintaining a core directional view.
The imperfect correlation between the two allows for more diversification and marginally lower risk.
A trader can use price differentials to extract an advantage in the market.
Let's use our perfectly negative correlation example from above with BTC/USD and EOS/USD.
On average (this is completely hypothetical, as I have not done the analysis for price movement on these currencies) let's say Bitcoin moves $10, and EOS moves $1.
Since they are perfectly negatively correlated, a price movement upwards would be equal to $9 (Bitcoin up $10, EOS down $1). This acts as a nice hedge for the trader. The downside, of course, is smaller profits in the event of a strong Bitcoin move.
Financial analysts often disagree how correlations should be utilized. This specific correlation (Pearson) has several disadvantages:
This approach is only valid for linear dependencies; straight-line relationships between two assets are not often observed
Correlation only captures the first two moments of the relationship.
A value of 0 does not necessarily mean a relationship is missing.
Oh yes...big time. Let's take a look at Ethereum Classic (ETC) and EOS (EOS):
You can see it ranged all the way from -0.40 to almost a perfect correlation at 0.95.
What does this mean?
It could mean a few things. First, events that impact a specific crypto but not another could wildly move the price against the comparison.
Second, we are still in a young industry, and movements have not come even close to settling. We are still in a period of great uncertainty surrounding cryptocurrencies.
Cryptocurrency correlation coefficients can be a very useful tool when assessing your next crypto investment.
The benefits of the analysis, such as helping to avoid canceling positions, diversification, and hedging are second-to-none.
I urge you to download the spreadsheet so you can do your own analysis. See how they change on an hourly, daily, weekly, and monthly basis. You are bound to uncover some tremendous insights that others in the market may not be seeing.
Cheers, and happy hunting!
Download the add-in: https://spreadstreet.io/tools/google-sheets-add-in
Help: https://spreadstreet.io/docs
First time install and login:
Bitfinex candles endpoint help: https://spreadstreet.io/knowledge-base/bitfinex-api-candles-endpoint/
10 Statistical Price Predictions for 10 Cryptocurrencies
High-Flyers and Shitcoins: What I Learned from Analyzing CoinMarketCap Data in Google Sheets
7 Smart Ethereum Price Prediction Methods for HODL’ers
