#!/usr/bin/env python
# coding: utf-8

# # Daily Exchange Rates per Euro #
# 
# Our focus in the guided part of the project will be on the exchange rate between the euro and the American dollar.
# 
# Daria Chemkaeva put together the data set and made it available on [Kaggle](https://www.kaggle.com/lsind18/euro-exchange-daily-rates-19992020) — the data source is the European Central Bank. Note that the dataset gets regular updates — we downloaded it on January 2021.

# In[138]:


import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt 
import datetime as dt
# Enables Jupyter to display graphs
get_ipython().run_line_magic('matplotlib', 'inline')
exchange_rates = pd.read_csv('euro-daily-hist_1999_2020.csv')



# In[139]:


#learn some basic facts about the dataset:
exchange_rates.info()


# In[140]:


exchange_rates.head()


# In[141]:


exchange_rates.tail()


# __Basic Facts__
# 
# - Rows 5699
# - Col 41
# - Null values: 0
# - Data types: object & float

# __Data Cleaning__

# In[142]:


#We rename the [US dollar ] and Period\Unit: columns to something easier to type — US_dollar and Time.
exchange_rates.rename(columns={'[US dollar ]': 'US_dollar','Period\\Unit:': 'Time'},inplace=True)

#We change the Time column to a datetime data type.
exchange_rates['Time'] = pd.to_datetime(exchange_rates['Time'])

#We sort the values by Time in ascending order.
exchange_rates.sort_values('Time', inplace=True)

#We reset the index (and drop the initial index).
exchange_rates.reset_index(drop=True, inplace=True)


# In[143]:


#Isolate the Time and the US_dollar columns. Assign them to a different variable named euro_to_dollar.
euro_to_dollar = exchange_rates[['Time','US_dollar']]

#Run the Series.value_counts() method on the US_dollar column, and see if you notice anything wrong.
euro_to_dollar['US_dollar'].value_counts()

#Drop all the rows where the - character appears in the US_dollar column(# - 62)
euro_to_dollar = euro_to_dollar[euro_to_dollar['US_dollar'] != '-']

#Convert the US_dollar column to a float data type.
euro_to_dollar["US_dollar"] = euro_to_dollar["US_dollar"].astype(float)
euro_to_dollar.head(20)


# 
# __Generated line plot to visualize the evolution of the euro-dollar exchange rate.__

# In[144]:


plt.style.use('default')
plt.plot(euro_to_dollar['Time'],
         euro_to_dollar['US_dollar'])
plt.show()


# __Rolling mean__

# In[145]:


#Calculate the rolling means for the US_dollar(30 days) column using a moving window of 30 days. Add the rolling means to a new column named rolling_mean.
#plt.style.use('default')
euro_to_dollar['rolling_mean'] = euro_to_dollar['US_dollar'].rolling(2).mean()


# In[146]:


#NOTE: FOR LOOP SYNTAX IS MORE CONSISE TO IMPLEMENT GRID

plt.figure(figsize=(10, 12))#controls size of grid

#1st subplot
plt.subplot(3, 2, 1)
plt.plot(euro_to_dollar['Time'],
         euro_to_dollar['US_dollar'])
plt.title('Orginal value',fontweight="bold")
    

#2nd
plt.subplot(3, 2, 2)
plt.plot(euro_to_dollar['Time'],euro_to_dollar['US_dollar'].rolling(7).mean())
plt.title('Rolling window 2 days',fontweight="bold")

#3rd 
plt.subplot(3, 2, 3)
plt.plot(euro_to_dollar['Time'],euro_to_dollar['US_dollar'].rolling(30).mean())
plt.title('rolling window 7 days',fontweight="bold")

#4rd
plt.subplot(3, 2, 4)
plt.plot(euro_to_dollar['Time'],euro_to_dollar['US_dollar'].rolling(50).mean())
plt.title('rolling window 50 days',fontweight="bold")

#5th
plt.subplot(3, 2,5)
plt.plot(euro_to_dollar['Time'],euro_to_dollar['US_dollar'].rolling(100).mean())
plt.title('rolling window 100 days', fontweight="bold")

#6th
plt.subplot(3, 2,6)
plt.plot(euro_to_dollar['Time'],euro_to_dollar['US_dollar'].rolling(365).mean())
plt.title('rolling mean 365 days',fontweight="bold")

plt.show()



# __Story telling__
# 
# We show how the euro-dollar rate changed during the 2007-2008 financial crisis. We can also show the data for 2006 and 2009 for comparison. We can use a line plot.

# __Project: Euro-USD rate peaked at 1.59 during 2007-2008 financial crisis.__

# In[147]:


#get dates for usd exchange rates 2006-2010
exchange_rate_06_10 = euro_to_dollar.copy()[(euro_to_dollar['Time'].dt.year >= 2006) & (euro_to_dollar['Time'].dt.year <= 2009)]

#get dates for usd exchange rates 2007-2009
financial_07_08 = euro_to_dollar.copy()[(euro_to_dollar['Time'].dt.year > 2007) & (euro_to_dollar['Time'].dt.year < 2009)]

#financial_07_08.describe()
#exchange_rate_06_10['Time']


# In[148]:


euro_to_dollar.columns


# In[151]:


#line graph to show changes
import matplotlib.style as style
#style.available - check available styles for graph
style.use('fivethirtyeight')#style of graph

#adding the graph
fig, ax = plt.subplots(figsize=(12, 8))#controls size of figure

#2006 - 10 finanacial rate

ax.plot(exchange_rate_06_10['Time'],
exchange_rate_06_10['rolling_mean'],color = 'g')

#2007- 8 financial crisis

ax.plot(financial_07_08['Time'],
financial_07_08['rolling_mean'],color = 'red')
    

#add title and suptitle

plt.suptitle('Euro-USD rate peaked at 1.59 during 2007-2008\'s financial crisis',
fontweight = 'bold',fontsize = 22)

plt.title('Euro-USD rate peaked between 2006 - 2010',y = 1
,x = 0.245)

#highlight peek exchange rate with vertical line
#ax.axvspan([exchange_rate_06_10(2007), exchange_rate_06_10(2007)],facecolor='#2ca02c')


#modify xticks and yticks


#add signature
#ax.text(-0.1, -1,
        #'©DATAQUEST     Source: P. Cortez et al.',
       # color = '#f0f0f0',
       # backgroundcolor = '#4d4d4d',
        #size=12)
           
plt.show()


# In[ ]: