How to Calculate Correlation Between Bitcoin & Ethereum in Python

July 4, 2022, 11:48 p.m.by CryptoDataDownload

In this article, we will calculate Pearson's correlation (often thought of as linear correlation) for Bitcoin (BTC) and Ethereum (ETH) using two different methods in Python. The timeseries data will come from the Gemini exchange page on our site; and we will use the daily interval (correlation can be measured at smaller time intervals as well). The two libraries we will use are: Pandas & SciStats. Interestingly, the SciStats method will also calculate a p-value for the relationship. The code is fully documented --> meaning each line so that you can easily follow the code logic! There are numerous resources available to learn more about correlation and the math behind it, so we will not cover those here. Code resources below

# CryptoDataDownload
# No Express Warranty
import ssl
import pandas as pd
import scipy.stats

# change the ssl state so we do not see an error for https:// content
ssl._create_default_https_context = ssl._create_unverified_context

def main():
    # first we will load the data and receive two lists of change metrics
    btc, eth = load_data_sets()

# next calculate correlation using Pandas
    calc_pandas_correlation(btc, eth)  # pass lists to function

# calculate correlation using scipy.stats
    calc_scipy_correlation(btc, eth)

def load_data_sets():
    """This function will load the BTC and ETH timeseries from the Gemini exchange and put them into Pandas
    dataframes. Then it will calculate a column for Percent Change using the closing values; and we will run correlation
    stats on the % chg column. Finally, function will return a list for BTC timeseries and one for ETH for the last 90
    days

"""
    # read the two data sets into dataframes and then return the dataframes as a Tuple
    btc_df = pd.read_csv("https://www.cryptodatadownload.com/cdd/Gemini_BTCUSD_d.csv", skiprows=1)
    eth_df = pd.read_csv("https://www.cryptodatadownload.com/cdd/Gemini_ETHUSD_d.csv", skiprows=1)

# calculate the pct change
    btc_df.sort_values('date', inplace=True)  # we sort the values so that oldest are first
    eth_df.sort_values('date', inplace=True)
    btc_df['pct_chg'] = btc_df['close'].pct_change(periods=1)
    eth_df['pct_chg'] = eth_df['close'].pct_change(periods=1)
    btc_chg_list = btc_df['pct_chg'].tolist() # convert the percent change column to a list
    eth_chg_list = eth_df['pct_chg'].tolist()  # convert the percent change column to a list

return btc_chg_list[-90:], eth_chg_list[-90:] # return only the last 90 days of timeseries

def calc_pandas_correlation(btc, eth):
    """The Pandas calculation can work on series themselves; so we will first convert our list object back into
    a Pandas series for BTC and ETH and then run the correlation function
    """
    btc_series = pd.Series(btc)
    eth_series = pd.Series(eth)
    corr = btc_series.corr(eth_series)
    print(f'Correlation using Pandas: {corr}')

def calc_scipy_correlation(btc, eth):
    """SciPy.stats correlation will calculate both the correlation and p-value.

"""
    corr, pval = scipy.stats.pearsonr(btc, eth) # returns tuple of corr, p-value
    print(f'Correlation using Scipy.stats: {corr}')

if __name__ == "__main__":
    """Simple script demonstrating how to calculate linear correlation of two cryptocurrency timeseries.
    In this case, we calculate the correlation between BTC and ETH over the last 90 days.
    """
    main()

Notice: Information contained herein is not and should not be construed as an offer, solicitation, or recommendation to buy or sell securities. The information has been obtained from sources we believe to be reliable; however no guarantee is made or implied with respect to its accuracy, timeliness, or completeness. Author does not own the any crypto currency discussed. The information and content are subject to change without notice. CryptoDataDownload and its affiliates do not provide investment, tax, legal or accounting advice.

This material has been prepared for informational purposes only and is the opinion of the author, and is not intended to provide, and should not be relied on for, investment, tax, legal, accounting advice. You should consult your own investment, tax, legal and accounting advisors before engaging in any transaction. All content published by CryptoDataDownload is not an endorsement whatsoever. CryptoDataDownload was not compensated to submit this article. Please also visit our Privacy policy; disclaimer; and terms and conditions page for further information.

About

CryptoDataDownload - Free market data and risk analytics tools for cryptocurrency research and analysis. Premium content & analytical offerings also available. Trusted by academics worldwide.

Navigation

Contacts

E-mail

support@cryptodatadownload.com
Have feedback or questions? Our team is here to help. Contact us for any inquiries or to report any inaccuracies.

How to Calculate Correlation Between Bitcoin & Ethereum in Python

Latest Posts

Follow Us

Notify me of new content

About

Navigation

Contacts