source: https://eurocoinpay.io/

COVID19 VS BITCOIN

COVID19 has changed the whole economy in the world. Many people has lost their jobs, their business, others, started as entrepreneurs, but through this analysis we can find out that a lot of people started to invest in cryptocurrencies.

It looks that COVID19 gave a huge push up on the value, reaching better values everyday through this pandemical situation.

Let’s start with the analysis.

Setup:

It is important to get an account in BINANCE or any other EXCHANGE website.

Now let’s get our canddlestick chart, this EXCHANGE allows us to get the last 500 periods, in this case, I decided to look for one day as an interval.

Now, lets get our COVID19 data:

Photo by Elizaveta Dushechkina on Unsplash

And start EDA of it:

Date appears as a “chr” datatype, so we must change it to a “date” type:

Look in to data to find inconsistent data, for example: cases < 0, deaths <0, popData2019 <0.

Our Covid19 dataset has many info, that is not relevant for us.

Start digging:

Looking for NA’s

121 NA’s? Where are them?

NA’s are on the column ‘countryterritoryCode’ and since those countries have a name, let’s just let it go and work with ‘countriesAndTerritories’.

For this study, global power economics will be used to compare. For example: USA, European Union, China and since some data look “weird” Japan has been added because is in the same region as China.

GROUPING

This article needs some graphics.

Next graph, show you Covid19 New Cases

dyRoller is added to this plot it is interactive on my repository. Link at the end.

Now, let’s add another graph with our groups.

China and Japan, are at the bottom of the graph, appearently with just a few cases.

We apply a smothing to the graph with ROLLMEAN:

This is getting interesting now, fitting our BTCUSDT data is the next objective. Since our Covid19 dataset starts on Dec 31 2019, we must have the same length on the BTCUSDT data, so is selected that same date and its chosen the column CLOSE and then changed the name.

Columns will be renamed.

Now, our groups will be joined:

Joining Covid19 data with Bitcoin value, is up next:

How does the data correlate?

Correlation within the values.

The correlation between the World New Cases vs BTCUSDT value is calculated as following:

Let’s plot.

Covid New Cases vs BTCUSDT.

BTCUSDT value through time:

BTCUSDT value since Covid19.

Pretty interesting info right?

FLEXDASHBOARD is used to build this study on a html dashboard. Screenshots are being added so you can look it up.

Check up the repository on: https://github.com/v-ibarra/Covid_vs_Bitcoin