[Section6] Data Visualization with Matplotlib

해당 내용은 Udemy 강의 'Pandas 및 Python을 이용한 데이터 분석:마스터 클래스'를 수강 후 정리한 내용입니다.

https://www.udemy.com/course/best-pandas-python

Section1

BASIC SINGLE LINE PLOT

# The plot method on Pandas Series and DataFrames is just a simple wrapper around plt.plot():

import matplotlib.pyplot as plt
import pandas as pd 

# Use matplotlib on the Pandas DataFrame to plot the data

investment_df.plot(x='Date', y='BTC-USD Price', label='Bitcoin Price', linewidth=3, figsize=(14, 6))

# label의 위치 수정
plt.ylabel('Price [$]')
plt.xlabel('Date')
plt.title('My First Data Visualization Code!')
plt.legend(loc='lower right')
plt.grid()

MINI CHALLENGE #1:

• Plot similar kind of graph for Ethereum instead
• Change the line color to red

investment_df.plot(x='Date', y='ETH-USD Price', label='Ethereum Price', linewidth=2, figsize=(14, 6), color='r')
plt.xlabel('Date')
plt.ylabel('Price [$]')
plt.title('MINI CHALLENGE #1')
plt.grid()
plt.legend(loc='lower right')
plt.show()

---

Section2

DOWNLOAD DATA DIRECTLY FROM YAHOO FINANCE

# yfinance library offers an easy, reliable and Pythonic way to download market data from Yahoo finance
! pip install yfinance
import yfinance as yf

# List all Ticker Symbols in the list below, let's start with one!

investments_list = ['BTC-USD']

# Specify the start and end dates

start = datetime.datetime(2010, 1, 1)
end = datetime.datetime(2021, 5, 16)

# Download the data from Yahoo Finance, make sure to reset index!

df = yf.download(investments_list, start=start, end=end)

# Reset index as follows:

df.reset_index(inplace=True)
df

MINI CHALLENGE #2:

• Use Yahoo Finance to download Ethereum data instead of BTC
• USe Yahoo Finance to download BTC, ETH, and LTC

# ETH data

investments_list = ['ETH-USD']

start = datetime.datetime(2010, 1, 1)
end = datetime.datetime(2021, 5, 16)

eth_df = yf.download(investments_list, start=start, end=end)
eth_df

# BTC, ETH and LTC

investments_list = ['BTC-USD','ETH-USD', 'LTC-USD']

start = datetime.datetime(2010, 1, 1)
end = datetime.datetime(2021, 5, 16)

finance_df = yf.download(investments_list, start=start, end=end)
finance_df

---

Section3

MULTIPLE PLOTS

investment_df.plot(x='Date', y=['BTC-USD Price', 'ETH-USD Price'], linewidth=2, figsize=(14,6))
plt.ylabel('Price')
plt.title('Crypto Prices')
plt.grid()

MINI CHALLENGE #3:

• Add Litecoin (LTC) to the list and plot similar kind of graph showing all three crypto currencies

investment_df.plot(x='Date', y=['BTC-USD Price', 'ETH-USD Price', 'LTC-USD Price'], linewidth=1, figsize=(14,6))

---

Section4

SUBPLOTS

investment_df.plot(x='Date', title='Crypto Prices', subplots=True, grid=True, figsize=(15,25))

MINI CHALLENGE #4:

• Try to set subplots = False and examine the output.

investment_df.plot(x='Date', title='Crypto Prices', subplots=False, grid=True, figsize=(15,25))

investment_df.plot(x='Date', y=['BTC-USD Price', 'ETH-USD Price', 'LTC-USD Price'], linewidth=1, figsize=(14,6))

결과와 동일

---

Section5

SCATTERPLOT

# Read daily return data using pandas

daily_return_df = pd.read_csv('crypto_daily_returns.csv')
daily_return_df

# Plot Daily returns of BTC vs. ETH

daily_return_df.plot.scatter('BTC', 'ETH', grid=True, figsize=(12,7))

MINI CHALLENGE #5:

• Plot the daily returns of BTC vs. LTC

# Plot Daily returns of BTC vs. LTC

daily_return_df.plot.scatter('BTC', 'LTC', grid=True, figsize=(12,7))

---

Section6

PIE CHART

# Define a dictionary with all crypto allocation in a portfolio
# Note that total summation = 100% 

my_dict = {'allocation %': [20, 55, 5, 17, 3]}
my_dict

crypto_df = pd.DataFrame(data=my_dict, index=['BTC', 'ETH', 'LTC', 'XRP', 'ADA'])
crypto_df

# Use matplotlib to plot a pie chart 

crypto_df.plot.pie(y='allocation %', figsize=(8,8))
plt.title('Crypto Portfolio Pie Chart')

pie chart의 경우 y의 인자 필요

MINI CHALLENGE #6:

• Assume that you became bullish on XRP and decided to allocate 60% of your assets in it. You also decided to equally divide the rest of your assets in other coins (BTC, LTC, ADA, and ETH). Change the allocations and plot the pie chart.
• Use 'explode' to increase the separation between XRP and the rest of the portfolio (External Research is Required)

my_dict = {'allocation %': [10, 10, 10, 60, 10]}

crypto_df = pd.DataFrame(data=my_dict, index=['BTC', 'ETH', 'LTC', 'XRP', 'ADA'])

crypto_df.plot.pie(y='allocation %')
plt.title('Crypto Portfolio Pie Chart')

explode = (0, 0, 0, 0.2, 0)

crypto_df.plot.pie(y='allocation %', explode=explode)
plt.title('Crypto Portfolio Pie Chart')

---

Section7

HISTOGRAMS

# A histogram represents data using bars with various heights
# Each bar groups numbers into specific ranges
# Taller bars show that more data falls within that specific range

daily_return_df = pd.read_csv('crypto_daily_returns.csv')
daily_return_df

# Find the statistical value of the distribution

mu = daily_return_df['BTC'].mean()
sigma = daily_return_df['BTC'].std()

# Visualization

plt.figure(figsize=(12, 6))
# alpha는 graph의 투명도
daily_return_df['BTC'].plot.hist(bins=50, alpha=0.5)

plt.grid()
plt.title('Histogram: mu=' + str(mu) + ', sigma=' + str(sigma))

MINI CHALLENGE #7:

• Plot the histogram for ETH returns using 30 bins with red color

# Find the statistical value of the distribution

mu = daily_return_df['ETH'].mean()
sigma = daily_return_df['ETH'].std()

# Visualization

plt.figure(figsize=(12, 6))
daily_return_df['ETH'].plot.hist(bins=30, alpha=0.5, color='r')

plt.grid()
plt.title('Histogram: mu=' + str(mu) + ', sigma=' + str(sigma))

---