Bitcoin, Altcoins and Finding Value with Linear Regression

Bitcoin has had a storming few weeks. And although good for cryptocurrency in general, has largely come at the expense of Altcoins, Bitcoin’s long suffering little brother. This is nicely demonstrated below, where as you can see Bitcoin has had another +20% week while Altcoins have lingered, with the average Altcoin on Binance ending last week almost unchanged.

So while we are waiting for this trend to reverse, we thought it would be interesting to see, statistically speaking, which of the Top 30 are currently offering the best value, both against Bitcoin and against their most correlated pairing.

So we gathered the last 6 months of price data for the current Top 30 coins, and using Sci-Kit Learn’s Linear Regression model compared each coin against Bitcoin and against each coins most correlated pair. And while there are many limitations with Linear Regression and many coins aren’t that well correlated with Bitcoin, the results are still interesting to see.

As you can see almost all Altcoins look undervalued compared to Bitcoin, either making these coins cheap, or Bitcoin expensive. It should also be noted that the correlation for many isn’t that strong, so these results need to be taken with a pinch of salt.

Here we have the table of each coin in comparison with it’s most correlated pair, with most pairs having a much better correlation one could perhaps take these results a bit more literally, but again this is just suggesting one is cheap or expensive in comparison with the other.

So in an attempt to draw some meaning from all of this one could argue that coins such as Bitcoin Cash, BAT or VeChain probably offer the best value using the study above, meaning compared to the rest of the market they look cheap. GL.