Innovative, Python, Rstudio, Tableau

Cryptocurrency pricing volatility, and predictive analysis

In this project, I gathered price data off a cryptocurrency exchange (web scraping using pyhon beautiful soup), and used Tableau and R for visuals and predictive analysis.

Web Scraping Cryptocurrency price

Packages

import requests
from bs4 import BeautifulSoup

Code

url = 'https://coinmarketcap.com/currencies/electroneum/'
response = requests.get(url)
html = response.content
#print(html)
soup = BeautifulSoup(html, 'html.parser')
price = soup.find(class_="priceValue")
print(price.text)

Tableau Visuals and predictive analytics

Unfortunately, I am not able to display my tableau workbook here. The visuals and dashboard work can be surmised from a review of the finished presentation at the bottom of the post.

R code and visuals

Code

"Quinton Carter" <-Sys.info()

install.packages("ggplot2")
library(ggplot2)

Crypto_Data <- read.csv(file.choose())
Crypto2 <- read.csv(file.choose())

str(Crypto_Data)
CData <- Crypto_Data
summary(CData)


CData$BTC_open_price <- as.factor(CData$BTC_open_price)
CData$BTC_high_price <- as.factor(CData$BTC_high_price)
CData$BTC_low_price <- as.factor(CData$BTC_low_price)
CData$BTC_close_price <- as.factor(CData$BTC_close_price)
CData$BTC_volume <- as.factor(CData$BTC_volume)
CData$BTC_Hi_Lo_range <- as.factor(CData$BTC_Hi_Lo_range)
CData$BTC_Daily_range <- as.factor(CData$BTC_Daily_range)
CData$ETH_open_price <- as.factor(CData$ETH_open_price)
CData$ETH_high_price <- as.factor(CData$ETH_high_price)
CData$ETH_low_price <- as.factor(CData$ETH_low_price)
CData$ETH_close_price <- as.factor(CData$ETH_close_price)
CData$ETH_volume <- as.factor(CData$ETH_volume)
CData$ETH_Hi_Lo_range <- as.factor(CData$ETH_Hi_Lo_range)
CData$ETH_Daily_range <- as.factor(CData$ETH_Daily_range)
CData$MANA_open_price <- as.factor(CData$MANA_open_price)
CData$MANA_high_price <- as.factor(CData$MANA_high_price)
CData$MANA_low_price <- as.factor(CData$MANA_low_price)
CData$MANA_close_price <- as.factor(CData$MANA_close_price)
CData$MANA_volume <- as.factor(CData$MANA_volume)
CData$MANA_Hi_Lo_range <- as.factor(CData$MANA_Hi_Lo_range)
CData$MANA_Daily_range <- as.factor(CData$MANA_Daily_range)

CData$BTC_AVG <- as.factor(CData$BTC_AVG)
CData$ETH_AVG <- as.factor(CData$ETH_AVG)
CData$MANA_AVG <- as.factor(CData$MANA_AVG)



ggplot(data = CData,aes(x=BTC_volume, y=WeekDay)) + 
  geom_bar(stat = "identity")


ggplot(aes(x = BTC_AVG, color = Month) ,data = CData) + 
  geom_density()

ggplot(CData, aes(x=Day,y=BTC_AVG))+
  geom_point(aes(color = factor(Year)))

ggplot(CData, aes(x=Day,y=BTC_AVG))+
  geom_point(aes(color = factor(WeekDay)))

                 
DVizBTC <- ggplot(data=CData, aes(x=BTC_AVG)) + 
  geom_histogram(aes(y=..density..),
                 col='black', 
                 fill='dodgerblue1', 
                 alpha=0.3) +
  geom_density(adjust=3)
print(DVizBTC + theme(plot.title=element_text(face="bold")) + ggtitle('Distribution AVG BTC'))

DVizETH <- ggplot(data=CData, aes(x=ETH_AVG)) + 
  geom_histogram(aes(y=..density..),
                 col='black', 
                 fill='dodgerblue1', 
                 alpha=0.3) +
  geom_density(adjust=3)
print(DVizETH + theme(plot.title=element_text(face="bold")) + ggtitle('Distribution AVG ETH'))


DVizMANA <- ggplot(data=CData, aes(x=MANA_AVG)) + 
  geom_histogram(aes(y=..density..),
                 col='black', 
                 fill='dodgerblue1', 
                 alpha=0.3) +
  geom_density(adjust=3)
print(DVizMANA + theme(plot.title=element_text(face="bold")) + ggtitle('Distribution AVG MANA'))


ggplot(data=CData, aes(x=Day,y=BTC_AVG,color=BTC_volume))+
  geom_point()+
  stat_smooth(method= "lm", col = "green")+
  geom_smooth()+
  ggtitle('BTC Preds')

ggplot(data=CData, aes(x=Month,y=ETH_AVG,color=ETH_volume))+
  geom_point()+
  stat_smooth(method= "lm", col = "green")+
  geom_smooth()+
  ggtitle('ETH Preds')

ggplot(data=CData, aes(x=Day,y=MANA_AVG,color=MANA_volume))+
  geom_point()+
  stat_smooth(method= "lm", col = "green")+
  geom_smooth()+
  ggtitle('MANA Preds')



corVizBTC <- select(CData, Date,Day,WeekDay,Month,Year,BTC_open_price,BTC_high_price,BTC_low_price,BTC_close_price,BTC_volume,BTC_Hi_Lo_range,BTC_Daily_range,BTC_AVG)
str(corVizBTC)
corVizBTC$Day <- as.numeric(corVizBTC$DAY)
corVizBTC$Year <- as.numeric(corVizBTC$Year)

numcolBTC <- sapply(corVizBTC,is.numeric)
pearsoncorBTC <- cor(corVizBTC[numcolBTC], use="complete.obs")
corrplot(pearsoncorBTC, "number")

corVizETH <- select(CData, Date,Day,WeekDay,Month,Year,ETH_open_price,ETH_high_price,ETH_low_price,ETH_close_price,ETH_volume,ETH_Hi_Lo_range,ETH_Daily_range,ETH_AVG)
str(corVizETH)
corVizETH$Day <- as.numeric(corVizETH$DAY)
corVizETH$Year <- as.numeric(corVizETH$Year)
numcolETH <- sapply(corVizETH,is.numeric)
pearsoncorETH <- cor(corVizETH[numcolETH], use="complete.obs")
corrplot(pearsoncorETH, "number")



corVizMANA <- select(CData, Date,Day,WeekDay,Month,Year,MANA_open_price,MANA_high_price,MANA_low_price,MANA_close_price,MANA_volume,MANA_Hi_Lo_range,MANA_Daily_range,MANA_AVG)
str(corVizMANA)
corVizMANA$Day <- as.numeric(corVizMANA$DAY)
corVizMANA$Year <- as.numeric(corVizMANA$Year)
numcolMANA <- sapply(corVizMANA,is.numeric)
pearsoncorMANA <- cor(corVizMANA[numcolMANA], use="complete.obs")
corrplot(pearsoncorMANA, "number")


corVizAll <- select(CData, Date,Day,WeekDay,Month,Year,BTC_open_price,BTC_high_price,BTC_low_price,BTC_close_price,BTC_volume,BTC_Hi_Lo_range,BTC_Daily_range,BTC_AVG,ETH_open_price,ETH_high_price,ETH_low_price,ETH_close_price,ETH_volume,ETH_Hi_Lo_range,ETH_Daily_range,ETH_AVG,MANA_open_price,MANA_high_price,MANA_low_price,MANA_close_price,MANA_volume,MANA_Hi_Lo_range,MANA_Daily_range,MANA_AVG)
str(corVizAll)
corVizAll$Day <- as.numeric(corVizAll$DAY)
corVizAll$Year <- as.numeric(corVizAll$Year)
numcolAll <- sapply(corVizAll,is.numeric)
pearsoncorAll <- cor(corVizAll[numcolAll], use="complete.obs")
corrplot(pearsoncorAll, "number")


#Crypto2 charts

Crypto2$BTC_Day_Up <- as.factor(Crypto2$BTC_Day_Up)
Crypto2$BTC_DOD_Up <- as.factor(Crypto2$BTC_DOD_Up)
Crypto2$ETH_Day_Up <- as.factor(Crypto2$ETH_Day_Up)
Crypto2$ETH_DOD_Up <- as.factor(Crypto2$ETH_DOD_Up)
Crypto2$MANA_Day_Up <- as.factor(Crypto2$MANA_Day_Up)
Crypto2$MANA_DOD_Up <- as.factor(Crypto2$MANA_DOD_Up)

Crypto2 <- filter(Crypto2, Crypto2$BTC_volume < )

plot(BTC_AVG ~ ETH_Day_Up, data=Crypto2)
plot(ETH_AVG ~ MANA_Day_Up, data=Crypto2)


ggplot(Crypto2, aes(x = BTC_volume, y = BTC_AVG)) +
  geom_point(aes(color = factor(BTC_Day_Up))) +
  stat_smooth(method = "lm", col = "blue")+
  labs(title="BTC Vol to AVG")

ggplot(Crypto2, aes(x = ETH_volume, y = ETH_AVG)) +
  geom_point(aes(color = factor(ETH_Day_Up))) +
  stat_smooth(method = "lm", col = "blue")+
  labs(title="ETH Vol to AVG")

ggplot(Crypto2, aes(x = MANA_volume, y = MANA_AVG)) +
  geom_point(aes(color = factor(MANA_Day_Up))) +
  stat_smooth(method = "lm", col = "blue")+
  labs(title="MANA Vol to AVG")


NAVGdist <- ggplot(data=Crypto2, aes(x=Crypto2$WeekDay, fill=BTC_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdist + theme(plot.title=element_text(face="bold")) + ggtitle('BTC WeekDay'))

#Monthly
NAVGdistBMonth <- ggplot(data=Crypto2, aes(x=Crypto2$Month, fill=BTC_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistBMonth + theme(plot.title=element_text(face="bold")) + ggtitle('BTC Monthly'))

NAVGdistEMonth <- ggplot(data=Crypto2, aes(x=Crypto2$Month, fill=ETH_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistEMonth + theme(plot.title=element_text(face="bold")) + ggtitle('ETH Monthly'))

NAVGdistMMonth <- ggplot(data=Crypto2, aes(x=Crypto2$Month, fill=MANA_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistMMonth + theme(plot.title=element_text(face="bold")) + ggtitle('MANA Monthly'))

#Year
NAVGdistBYear <- ggplot(data=Crypto2, aes(x=Crypto2$Year, fill=BTC_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistBYear + theme(plot.title=element_text(face="bold")) + ggtitle('BTC Year'))

NAVGdistEYear <- ggplot(data=Crypto2, aes(x=Crypto2$Year, fill=ETH_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistEYear + theme(plot.title=element_text(face="bold")) + ggtitle('ETH Year'))

NAVGdistMYear <- ggplot(data=Crypto2, aes(x=Crypto2$Year, fill=MANA_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistMYear + theme(plot.title=element_text(face="bold")) + ggtitle('MANA Year'))


#WeekDay
NAVGdistBWeekDay <- ggplot(data=Crypto2, aes(x=Crypto2$WeekDay, fill=BTC_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistBWeekDay + theme(plot.title=element_text(face="bold")) + ggtitle('BTC WeekDay'))

NAVGdistEWeekDay <- ggplot(data=Crypto2, aes(x=Crypto2$WeekDay, fill=ETH_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistEWeekDay + theme(plot.title=element_text(face="bold")) + ggtitle('ETH WeekDay'))

NAVGdistMWeekDay <- ggplot(data=Crypto2, aes(x=Crypto2$WeekDay, fill=MANA_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistMWeekDay + theme(plot.title=element_text(face="bold")) + ggtitle('MANA WeekDay'))


#Day
NAVGdistBDay <- ggplot(data=Crypto2, aes(x=Crypto2$Day, fill=BTC_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistBDay + theme(plot.title=element_text(face="bold")) + ggtitle('BTC Day'))

NAVGdistEDay <- ggplot(data=Crypto2, aes(x=Crypto2$Day, fill=ETH_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistEDay + theme(plot.title=element_text(face="bold")) + ggtitle('ETH Day'))

NAVGdistMDay <- ggplot(data=Crypto2, aes(x=Crypto2$Day, fill=MANA_Day_Up)) + 
  geom_bar(aes(y = (..count..)/sum(..count..)), position='stack', alpha=0.5) + scale_y_continuous(labels=scales::percent)
print(NAVGdistMDay + theme(plot.title=element_text(face="bold")) + ggtitle('MANA Day'))

Visuals output aligned with their code

data-viz-from-r Download

Final presentation

Recording

Written analysis

final-crypto-compare Download

Slide deck

final-presentation-video-only Download