In order to provide a methodology to detect bubbles in time series, Phillips, Wu, and Yu [23] proposed the supremum augmented Dickey-Fuller (SADF) test. This applies a series of right-tailed unit root tests to expanding windows of a time series (with a fixed start date), defined by where r2 is the final data point to be considered in each window, starting at r0 which is a fraction representing the smallest allowed window size and expanding to 1(the complete data set).
Using data from a prominent blockchain research firm, Coinmetrics.io, we explore the relationship between prices, computing power (measured by the hashrate), and network (measured by the number of unique active users). We focus on five major mineable cryptocurrencies (i.e. Bitcoin, Ethereum, Monero, Litecoin, and Dash). In some of our test, we expand our data to include an additional 33 cryptocurrencies. Our sample period spans from 7 August 2015 to 25 January 2019.

A hardware "cold" wallet for safely storing and using your crypto. Meant as a better replacement with more features than competitors, and military grade encryption.
Google Trends returns data with different granularity depending on the historical time interval queried: daily search volumes are returned for queries under 90 days and weekly search volumes for queries of length over 90 days. However it is possible to reconstruct daily data for long time intervals using a combination of daily and weekly data and the method described and validated by [4]: Daily data is retrieved in buckets of under 90 days, and weekly data is also retrieved for the complete time interval of interest. Then using the daily data, the percentage change of each day in a week from the first day of the week is calculated; these percentage changes are then applied to the weekly data to build a daily time series over a longer period.