Cryptocurrency Analysis with Python — Log Returns

Roman Orac
Towards Data Science
4 min readOct 18, 2019

--

Photo by André François McKenzie on Unsplash

In the previous post, we analyzed raw price changes of cryptocurrencies. The problem with that approach is that prices of different cryptocurrencies are not normalized and we cannot use comparable metrics.

In this post, we describe the benefits of using log returns for analysis of price changes. You can download this Jupyter Notebook and the data.

--

--