international_comparison.py

   1 #!/usr/bin/env python
   2 # coding: utf-8
   3 # %%
   4 # Data from [European Centre for Disease Prevention and Control](https://www.ecdc.europa.eu/en/publications-data/download-todays-data-geographic-distribution-covid-19-cases-worldwide)
   5
   6 # %%
   7 import itertools
   8 import collections
   9 import json
  10 import pandas as pd
  11 import numpy as np
  12 from scipy.stats import gmean
  13 import datetime
  14
  15 from sqlalchemy.types import Integer, Text, String, DateTime, Date, Float
  16 from sqlalchemy import create_engine
  17
  18 import matplotlib as mpl
  19 import matplotlib.pyplot as plt
  20 plt.ioff()
  21 # %matplotlib inline
  22
  23
  24 # %%
  25 connection_string = 'postgresql://covid:3NbjJTkT63@localhost/covid'
  26
  27
  28 # %%
  29 engine = create_engine(connection_string)
  30
  31
  32 # %%
  33 DEATH_COUNT_THRESHOLD = 10
  34 COUNTRIES_CORE = tuple(sorted('ITA DEU GBR ESP IRL FRA BEL'.split()))
  35 # COUNTRIES_NORDIC = tuple('SE NO DK FI UK'.split())
  36 COUNTRIES_FRIENDS = tuple('ITA GBR ESP BEL SVN MEX'.split())
  37 # COUNTRIES_FRIENDS = 'IT UK ES BE SI PT'.split()
  38
  39 # COUNTRIES_AMERICAS = ('AG', 'AR', 'AW', 'BS', 'BB', 'BZ', 'BM', 'BO', 'BR', 'VG', 'KY', # excluding Canada and USA
  40 #        'CL', 'CO', 'CR', 'CU', 'CW', 'DM', 'DO', 'EC', 'SV', 'GL', 'GD', 'GT',
  41 #        'GY', 'HT', 'HN', 'JM', 'MX', 'MS', 'NI', 'PA', 'PY', 'PE', 'PR', 'KN',
  42 #        'LC', 'VC', 'SX', 'SR', 'TT', 'TC', 'VI', 'UY', 'VE')
  43 # COUNTRIES_OF_INTEREST = tuple(set(COUNTRIES_CORE + COUNTRIES_FRIENDS))
  44 # COUNTRIES_ALL = tuple(set(COUNTRIES_CORE + COUNTRIES_FRIENDS + COUNTRIES_NORDIC + COUNTRIES_AMERICAS))
  45
  46
  47 # %%
  48 query_string = f'''select date, country_code, deaths_weekly, culm_deaths
  49 from weekly_cases
  50 where country_code in {COUNTRIES_CORE}
  51 order by date, country_code'''
  52
  53 country_data = pd.read_sql_query(query_string,
  54                   engine,
  55                   index_col = 'date',
  56                   parse_dates = ['date']
  57                  )
  58
  59
  60 # %%
  61 deaths_culm = country_data.pivot(columns='country_code', values='culm_deaths')
  62
  63
  64 # %%
  65 # country_data
  66
  67 # %%
  68 # deaths_culm
  69
  70 # %%
  71 ax = deaths_culm.loc['2020-03-15':].plot(figsize=(10, 6), title="Total deaths, linear")
  72 ax.set_xlabel(f"Date")
  73 for c in COUNTRIES_CORE:
  74     lvi = deaths_culm[c].last_valid_index()
  75     ax.text(x = lvi + pd.Timedelta(days=1), y = deaths_culm[c][lvi], s = f"{c}: {deaths_culm[c][lvi]:.0f}")
  76 plt.savefig('covid_deaths_total_linear.png')
  77
  78
  79 # %%
  80 deaths_weekly = country_data.pivot(columns='country_code', values='deaths_weekly')
  81
  82
  83 # %%
  84 ax = deaths_weekly.loc['2020-03-01':, COUNTRIES_CORE].plot(figsize=(10, 6), title="Deaths per week")
  85 ax.set_xlabel('Date')
  86 for c in COUNTRIES_CORE:
  87     lvi = deaths_weekly[c].last_valid_index()
  88     ax.text(x = lvi + pd.Timedelta(days=1), y = deaths_weekly[c][lvi], s = c)
  89 plt.savefig('covid_deaths_per_week.png')
  90
  91
  92 # %%
  93 ax = deaths_weekly.iloc[-6:].plot(figsize=(10, 6), title="Deaths per week, last 6 weeks")#, ylim=(-10, 100))
  94 ax.set_xlabel("Date")
  95
  96 text_x_pos = deaths_weekly.last_valid_index() + pd.Timedelta(days=0.5)
  97
  98 for c in COUNTRIES_CORE:
  99     lvi = deaths_weekly[c].last_valid_index()
 100 #     if c != 'ES':
 101     ax.text(x = text_x_pos, y = deaths_weekly[c][lvi], s = f"{c}: {deaths_weekly[c][lvi]:.0f}")
 102 plt.savefig('deaths_by_date_last_6_weeks.png')
 103
 104
 105 # %%
 106
 107
 108
 109