case1.py

import math
import numpy as np
import pandas as pd
import plotly.express as px

# Metadata
data = pd.read_csv('data/Case1.csv')
print(data.info())
print(data.describe())
data['Date'] = pd.to_datetime(data['Date'])

'''
1. Display the long-term performance of the equity markets 
'''

data_melted = data.melt(id_vars=['Date'], value_vars=data.columns[1:], var_name='Index', value_name='Value')
fig = px.line(data_melted, x='Date', y='Value', color='Index', title='Long-term performance of the equity markets', log_y=True)
# fig.show()

'''
2. Calculate mean returns and volatilities over the total period
'''

returns = data.copy()
for index in data.columns[1:]:
    for i in range(1, len(data[index])):
        returns.loc[i, index] = np.log(data.loc[i, index] / data.loc[i - 1, index])

# First row needs to be removed as it is 100% return
returns = returns.drop(0)

print(returns.info())
print(returns.head())

# Mean and SD (volatility) of the returns
results_list = []
for index in returns.columns[1:]:
    mean_return = math.exp(returns[index].mean()*12)-1
    std_return = returns[index].std()*math.sqrt(12)
    results_list.append({'Index': index, 'Mean Return p.a.': mean_return, 'Volatility': std_return})

results = pd.DataFrame(results_list)
results_melted = results.melt(id_vars='Index', value_vars=['Mean Return p.a.', 'Volatility'], var_name='Metric', value_name='Value')
fig = px.bar(results_melted, x='Index', y='Value', color='Metric', barmode='group', title='Mean returns p.a. and volatilities')
# fig.show()

'''
3. Calculate yearly returns for the stock markets
'''

returns['Year'] = returns['Date'].dt.year
grouped_returns = returns.drop(columns=['Date']).groupby('Year').sum().reset_index()

fig = px.line(grouped_returns, x='Year', y=grouped_returns.columns[1:], title='Yearly returns for the stock markets')
# fig.show()

'''
4. Calculate the correlations between the markets over the period 1993-2017
'''

correlation_matrix = returns.drop(columns=['Date', 'Year']).corr()
correlation_matrix.loc['Mean'] = correlation_matrix.mean(axis=1)

fig = px.imshow(correlation_matrix, text_auto=True, title='Correlation Matrix (full)', color_continuous_scale='RdBu_r')
# fig.show()

'''
5. Calculate the correlations for five sub-periods and analyze the correlation regimes
– 1993-1997
– 1998-2002
– 2003-2007
– 2008-2012
– 2013-2017
'''

correlation_matrix_avg = pd.DataFrame()

time_delta = 1
for year in range(1993, 2018, time_delta):
    returns_sub = returns[returns['Year'].between(year, year+time_delta)]
    correlation_matrix = returns_sub.drop(columns=['Date', 'Year']).corr()
    correlation_matrix_avg[f'{year}-{year + time_delta}'] = correlation_matrix.mean()

    fig = px.imshow(correlation_matrix, text_auto=True, title=f'Correlation Matrix ({year}-{year + time_delta})', color_continuous_scale='RdBu_r', zmin=0, zmax=1)
    fig.show()

fig = px.imshow(correlation_matrix_avg, text_auto=True, title='Average correlation over sub-periods', color_continuous_scale='RdBu_r', zmin=0, zmax=1)
# fig.show()