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('IT DE UK ES IE FR BE'.split()))
  35 COUNTRIES_NORDIC = tuple('SE NO DK FI UK'.split())
  36 COUNTRIES_FRIENDS = tuple('IT UK ES BE SI MX'.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 report_date, geo_id, deaths_weekly, culm_deaths
  49 from weekly_cases
  50 where geo_id in {COUNTRIES_CORE}
  51 order by report_date, geo_id'''
  52
  53 country_data = pd.read_sql_query(query_string,
  54                   engine,
  55                   index_col = 'report_date',
  56                   parse_dates = ['report_date']
  57                  )
  58
  59
  60 # %%
  61 deaths_culm = country_data.pivot(columns='geo_id', values='culm_deaths')
  62
  63
  64 # %%
  65 ax = deaths_culm.loc['2020-03-15':].plot(figsize=(10, 6), title="Total deaths, linear")
  66 ax.set_xlabel(f"Date")
  67 for c in COUNTRIES_CORE:
  68     lvi = deaths_culm[c].last_valid_index()
  69     ax.text(x = lvi + pd.Timedelta(days=1), y = deaths_culm[c][lvi], s = f"{c}: {deaths_culm[c][lvi]:.0f}")
  70 plt.savefig('covid_deaths_total_linear.png')
  71
  72
  73 # %%
  74 deaths_weekly = country_data.pivot(columns='geo_id', values='deaths_weekly')
  75
  76
  77 # %%
  78 ax = deaths_weekly.loc['2020-03-01':, COUNTRIES_CORE].plot(figsize=(10, 6), title="Deaths per week")
  79 ax.set_xlabel('Date')
  80 for c in COUNTRIES_CORE:
  81     lvi = deaths_weekly[c].last_valid_index()
  82     ax.text(x = lvi + pd.Timedelta(days=1), y = deaths_weekly[c][lvi], s = c)
  83 plt.savefig('covid_deaths_per_week.png')
  84
  85
  86 # %%
  87 ax = deaths_weekly.iloc[-6:].plot(figsize=(10, 6), title="Deaths per week")#, ylim=(-10, 100))
  88 ax.set_xlabel("Date")
  89
  90 text_x_pos = deaths_weekly.last_valid_index() + pd.Timedelta(days=0.5)
  91
  92 for c in COUNTRIES_CORE:
  93     lvi = deaths_weekly[c].last_valid_index()
  94 #     if c != 'ES':
  95     ax.text(x = text_x_pos, y = deaths_weekly[c][lvi], s = f"{c}: {deaths_weekly[c][lvi]:.0f}")
  96 plt.savefig('deaths_by_date_last_6_weeks.png')
  97
  98
  99 # %%
 100
 101
 102
 103