Exploring Data In the Cryptocurrency Market App

Posted on Jun 12, 2021
The skills I demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.

Link to Shiny App | Author LinkedIn Page


     My interest in the subject of cryptocurrency and the pricing of coins arose from me seeing how prominently cryptocurrencies have been featured in the news over the past year. Earlier this year, I downloaded the app Coinbase (which allows user to buy and sell cryptocurrencies) to try and get a better sense of the cryptocurrency market data overall. But, there are quite a few crypto coins that users can buy/sell, and although the Coinbase UI is pretty good for looking at the price of a specific coin over the past 1 hour, 1 day, 1 week, 1 month, and 1 year, this view does not necessarily lend itself to giving the most complete picture of historical trends or trends across cryptocurrencies.

Overview & Research Questions

     The dataset that I found came from Kaggle.com and consisted of about 35k observations of the following: A cryptocurrency’s Name and Symbol, the date, the associated High Price and Low Price of the coin on that date, Opening Price and Closing Price for the Day, trading volume, and total market cap of the coin. On Coinbase, there is an option to set up recurring investment schedules, where you can pick a specific day of the week, every week, to buy a certain amount of a crypto coin. So, with the data and Coinbase in mind, I had the following questions: 
     1. Is there a historically ‘best’ schedule for weekly/monthly/quarterly/yearly investments that you can set, regardless of the specific week/month/quarter/year?
     2. Is there a tailored schedule that takes into account the specific month or quarter that differs from the best schedule that does not account for specific months? 
     3. Are there correlations from smaller coins to the more popular ones like Bitcoin?
     4. How do cryptocurrencies correlate to other indices, such as the NYSE or Nasdaq composites?
     I am approaching this app with the assumption that the user for this app would be someone looking to buy or sell cryptocurrency, but does not know how to approach the timing of purchases or sales, or possibly even know how cryptocurrency trading may differ from more  traditional investments.

Data Analysis

     The first step in my analysis was to bring in date and calendar attributes to aggregate across—day of week, week, month, quarter, and year. I was mainly interested in the change in price from opening to closing, so I focused on the percent change in daily price, which also normalized the data so that the metric could be easily compared across crypto coins (e.g. Bitcoin’s price as of Friday 6/4 was ~$37k, while Cardano’s was ~$1.75). From here I essentially split the analysis two ways.
     In the first split, I aggregated days of the week, days of the month, days of the quarter, and days of the year to find, based on history, the best time to schedule a recurring investment for various periods. Here, I was most interested in the maximum and minimum return days, so I included those in the data tables below each corresponding graph in the app. 
     In the app, this data is represented in the 'Dailies Aggregated' tab. Users can select from a dropdown menu the Cryptocurrency that they are interested in, and will then see a graph with historical performance by weekday, day of month, quarter, and year. Since there is no option to schedule investments on different days depending on what month you want to invest, for example, this part of the analysis is irrespective of specificities in the time series.

Returns by Day of Week

Exploring Data In the Cryptocurrency Market App

Return by Day of Month

Exploring Data In the Cryptocurrency Market App

Returns by Quarter

Exploring Data In the Cryptocurrency Market App

Returns by Day of Year


Daily Means Data

With the second split of the analysis, for each of the cryptocurrencies (starting with when they began to be tracked in the data) I took into account the specificities of the period to come up with the best historical performance of that period. For example, I wanted to find the best and worst day for returns for Bitcoin in the month of January across all years. Although following this schedule, you could not use the recurring investment option on Coinbase, I think that having more background as to how months differ in pricing cycles would be useful background information.
     This information is in the 'Returns' tab of the app, along with the associated highest/lowest day by month tables below the graph. 


   Next, I was very curious about correlations—between cryptocurrencies themselves, and also to the US Dollar, and then to the NYSE and Nasdaq composites. Looking just at the correlations between cryptocurrencies, I was not too surprised to see that most cryptocurrencies I looked at were fairly well correlated (>.5) to Bitcoin and Ethereum, two of the bigger crypto coins by market cap. 
Correlation Plot
     Looking at correlation between these crypto’s and the USD Coin, which is backed by the US Dollar was also very interesting—almost all of the cryptocurrencies were slightly negatively correlated to the USD Coin, suggesting investors may look to hedge their US dollars with crypto coins. Polkadot coin was the most negatively correlated to USD coin, but the correlation there was still above (-.4).
     And then finally, I dropped in the NYSE and Nasdaq composite returns for the same period to get another correlation comparison. Nearly all crypto’s were slightly positively correlated to the NYSE and Nasdaq, with USD Coin and Uniswap slightly negatively correlated to the stock indices, suggesting that the cryptocurrency market and stock markets tend to move with one another.

Data Results and Conclusion

     As a result of this analysis, I am coming away with a simple and clear tool users can use to examine historic pricing trends of cryptocurrencies for the purpose of scheduling investment with or without the ,  Additionally, I feel like I now know a bit more about the hedging rationalities underlying the cryptocurrency markets and how I could hedge when comparing to the US Dollar, and NYSE composite, and the Nasdaq composite. 


Nothing contained in this app or post should be construed as investment advice.     

Further Work

     I would love to keep updating this, adding new coins and updated daily information to this analysis, as well as adding additional fiat currencies and stock indicators, as well as adding on comparisons between these cryptocurrencies over the same periods.

About Author

Alexander Pinkerton

Extensive background leading merchandise planning, allocations, and buying functions, excited about the intersection of Data Science and consumer retail to drive the future of smarter decision making. NYC Data Science Academy Student Cohort 025 WashU '12 Bachelors of...
View all posts by Alexander Pinkerton >

Related Articles

Leave a Comment

No comments found.

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI