{
"cells": [
{
"cell_type": "code",
"execution_count": 127,
"metadata": {},
"outputs": [],
"source": [
"import itertools\n",
"import collections\n",
"import pandas as pd\n",
"import numpy as np\n",
"from scipy.stats import gmean\n",
"\n",
"import matplotlib as mpl\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" publishedweek1620201.xlsx publishedweek522018withupdatedrespiratoryrow.xls\n",
" publishedweek172020.csv publishedweek522019.xls\n",
" publishedweek172020.ods 'Weekly_Deaths - Historical_NI.xls'\n",
" publishedweek172020.xlsx Weekly_Deaths_NI_2020.xls\n",
" publishedweek522016.xls weekly-march-20-scotland.zip\n",
" publishedweek522017.xls\n"
]
}
],
"source": [
"!ls uk-deaths-data"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [],
"source": [
"raw_data_2020 = pd.read_csv('uk-deaths-data/publishedweek172020.csv', \n",
" parse_dates=[1], dayfirst=True,\n",
" index_col=0,\n",
" header=[0, 1])"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
" \n",
" Week number | \n",
" Week ended | \n",
" Total deaths, all ages | \n",
" Total deaths: average of corresponding week over the previous 5 years 1 (England and Wales) | \n",
" Total deaths: average of corresponding week over the previous 5 years 1 (England) | \n",
" Total deaths: average of corresponding week over the previous 5 years (Wales) | \n",
" Unnamed: 6_level_0 | \n",
" Unnamed: 7_level_0 | \n",
" Unnamed: 8_level_0 | \n",
" Persons | \n",
" Deaths by age group | \n",
" ... | \n",
" E12000001 | \n",
" E12000002 | \n",
" E12000003 | \n",
" E12000004 | \n",
" E12000005 | \n",
" E12000006 | \n",
" E12000007 | \n",
" E12000008 | \n",
" E12000009 | \n",
" W92000004 | \n",
"
\n",
" \n",
" | \n",
" Unnamed: 1_level_1 | \n",
" Unnamed: 2_level_1 | \n",
" Unnamed: 3_level_1 | \n",
" Unnamed: 4_level_1 | \n",
" Unnamed: 5_level_1 | \n",
" Deaths by cause | \n",
" Deaths where the underlying cause was respiratory disease (ICD-10 J00-J99) | \n",
" Deaths where COVID-19 was mentioned on the death certificate (ICD-10 U07.1 and U07.2) | \n",
" Unnamed: 9_level_1 | \n",
" <1 | \n",
" ... | \n",
" North East | \n",
" North West | \n",
" Yorkshire and The Humber | \n",
" East Midlands | \n",
" West Midlands | \n",
" East | \n",
" London | \n",
" South East | \n",
" South West | \n",
" Wales | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 2020-01-03 | \n",
" 12254 | \n",
" 12175 | \n",
" 11412 | \n",
" 756 | \n",
" NaN | \n",
" 2141 | \n",
" 0 | \n",
" NaN | \n",
" 48 | \n",
" ... | \n",
" 673 | \n",
" 1806 | \n",
" 1240 | \n",
" 1060 | \n",
" 1349 | \n",
" 1162 | \n",
" 1113 | \n",
" 1814 | \n",
" 1225 | \n",
" 787 | \n",
"
\n",
" \n",
" 2 | \n",
" 2020-01-10 | \n",
" 14058 | \n",
" 13822 | \n",
" 12933 | \n",
" 856 | \n",
" NaN | \n",
" 2477 | \n",
" 0 | \n",
" NaN | \n",
" 50 | \n",
" ... | \n",
" 707 | \n",
" 1932 | \n",
" 1339 | \n",
" 1195 | \n",
" 1450 | \n",
" 1573 | \n",
" 1272 | \n",
" 2132 | \n",
" 1487 | \n",
" 939 | \n",
"
\n",
" \n",
" 3 | \n",
" 2020-01-17 | \n",
" 12990 | \n",
" 13216 | \n",
" 12370 | \n",
" 812 | \n",
" NaN | \n",
" 2189 | \n",
" 0 | \n",
" NaN | \n",
" 69 | \n",
" ... | \n",
" 647 | \n",
" 1696 | \n",
" 1278 | \n",
" 1106 | \n",
" 1407 | \n",
" 1457 | \n",
" 1073 | \n",
" 2064 | \n",
" 1466 | \n",
" 767 | \n",
"
\n",
" \n",
" 4 | \n",
" 2020-01-24 | \n",
" 11856 | \n",
" 12760 | \n",
" 11933 | \n",
" 802 | \n",
" NaN | \n",
" 1893 | \n",
" 0 | \n",
" NaN | \n",
" 53 | \n",
" ... | \n",
" 612 | \n",
" 1529 | \n",
" 1187 | \n",
" 1024 | \n",
" 1231 | \n",
" 1410 | \n",
" 1028 | \n",
" 1833 | \n",
" 1253 | \n",
" 723 | \n",
"
\n",
" \n",
" 5 | \n",
" 2020-01-31 | \n",
" 11612 | \n",
" 12206 | \n",
" 11419 | \n",
" 760 | \n",
" NaN | \n",
" 1746 | \n",
" 0 | \n",
" NaN | \n",
" 50 | \n",
" ... | \n",
" 561 | \n",
" 1461 | \n",
" 1136 | \n",
" 1015 | \n",
" 1262 | \n",
" 1286 | \n",
" 1092 | \n",
" 1820 | \n",
" 1233 | \n",
" 727 | \n",
"
\n",
" \n",
"
\n",
"
5 rows × 84 columns
\n",
"
"
],
"text/plain": [
"Week number Week ended Total deaths, all ages \\\n",
" Unnamed: 1_level_1 Unnamed: 2_level_1 \n",
"1 2020-01-03 12254 \n",
"2 2020-01-10 14058 \n",
"3 2020-01-17 12990 \n",
"4 2020-01-24 11856 \n",
"5 2020-01-31 11612 \n",
"\n",
"Week number Total deaths: average of corresponding week over the previous 5 years 1 (England and Wales) \\\n",
" Unnamed: 3_level_1 \n",
"1 12175 \n",
"2 13822 \n",
"3 13216 \n",
"4 12760 \n",
"5 12206 \n",
"\n",
"Week number Total deaths: average of corresponding week over the previous 5 years 1 (England) \\\n",
" Unnamed: 4_level_1 \n",
"1 11412 \n",
"2 12933 \n",
"3 12370 \n",
"4 11933 \n",
"5 11419 \n",
"\n",
"Week number Total deaths: average of corresponding week over the previous 5 years (Wales) \\\n",
" Unnamed: 5_level_1 \n",
"1 756 \n",
"2 856 \n",
"3 812 \n",
"4 802 \n",
"5 760 \n",
"\n",
"Week number Unnamed: 6_level_0 \\\n",
" Deaths by cause \n",
"1 NaN \n",
"2 NaN \n",
"3 NaN \n",
"4 NaN \n",
"5 NaN \n",
"\n",
"Week number Unnamed: 7_level_0 \\\n",
" Deaths where the underlying cause was respiratory disease (ICD-10 J00-J99) \n",
"1 2141 \n",
"2 2477 \n",
"3 2189 \n",
"4 1893 \n",
"5 1746 \n",
"\n",
"Week number Unnamed: 8_level_0 \\\n",
" Deaths where COVID-19 was mentioned on the death certificate (ICD-10 U07.1 and U07.2) \n",
"1 0 \n",
"2 0 \n",
"3 0 \n",
"4 0 \n",
"5 0 \n",
"\n",
"Week number Persons Deaths by age group ... E12000001 E12000002 \\\n",
" Unnamed: 9_level_1 <1 ... North East North West \n",
"1 NaN 48 ... 673 1806 \n",
"2 NaN 50 ... 707 1932 \n",
"3 NaN 69 ... 647 1696 \n",
"4 NaN 53 ... 612 1529 \n",
"5 NaN 50 ... 561 1461 \n",
"\n",
"Week number E12000003 E12000004 E12000005 E12000006 \\\n",
" Yorkshire and The Humber East Midlands West Midlands East \n",
"1 1240 1060 1349 1162 \n",
"2 1339 1195 1450 1573 \n",
"3 1278 1106 1407 1457 \n",
"4 1187 1024 1231 1410 \n",
"5 1136 1015 1262 1286 \n",
"\n",
"Week number E12000007 E12000008 E12000009 W92000004 \n",
" London South East South West Wales \n",
"1 1113 1814 1225 787 \n",
"2 1272 2132 1487 939 \n",
"3 1073 2064 1466 767 \n",
"4 1028 1833 1253 723 \n",
"5 1092 1820 1233 727 \n",
"\n",
"[5 rows x 84 columns]"
]
},
"execution_count": 50,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"raw_data_2020.head()"
]
},
{
"cell_type": "code",
"execution_count": 78,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" Week number | \n",
" Week ended | \n",
" Total deaths, all ages | \n",
" Total deaths: average of corresponding | \n",
" Unnamed: 4_level_0 | \n",
" Unnamed: 5_level_0 | \n",
" Persons | \n",
" Unnamed: 7_level_0 | \n",
" Unnamed: 8_level_0 | \n",
" Unnamed: 9_level_0 | \n",
" Unnamed: 10_level_0 | \n",
" ... | \n",
" E12000001 | \n",
" E12000002 | \n",
" E12000003 | \n",
" E12000004 | \n",
" E12000005 | \n",
" E12000006 | \n",
" E12000007 | \n",
" E12000008 | \n",
" E12000009 | \n",
" W92000004 | \n",
"
\n",
" \n",
" | \n",
" Unnamed: 1_level_1 | \n",
" Unnamed: 2_level_1 | \n",
" Unnamed: 3_level_1 | \n",
" Deaths by underlying cause | \n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) | \n",
" Deaths by age group | \n",
" Under 1 year | \n",
" 01-14 | \n",
" 15-44 | \n",
" 45-64 | \n",
" ... | \n",
" North East | \n",
" North West | \n",
" Yorkshire and The Humber | \n",
" East Midlands | \n",
" West Midlands | \n",
" East | \n",
" London | \n",
" South East | \n",
" South West | \n",
" Wales | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 2019-01-04 | \n",
" 10955 | \n",
" 12273 | \n",
" NaN | \n",
" 1736 | \n",
" NaN | \n",
" 43 | \n",
" 15 | \n",
" 215 | \n",
" 1199 | \n",
" ... | \n",
" 574 | \n",
" 1456 | \n",
" 1148 | \n",
" 964 | \n",
" 1180 | \n",
" 1136 | \n",
" 1021 | \n",
" 1586 | \n",
" 1136 | \n",
" 718 | \n",
"
\n",
" \n",
" 2 | \n",
" 2019-01-11 | \n",
" 12609 | \n",
" 13670 | \n",
" NaN | \n",
" 2214 | \n",
" NaN | \n",
" 50 | \n",
" 20 | \n",
" 280 | \n",
" 1419 | \n",
" ... | \n",
" 643 | \n",
" 1612 | \n",
" 1279 | \n",
" 1089 | \n",
" 1299 | \n",
" 1386 | \n",
" 1137 | \n",
" 1927 | \n",
" 1394 | \n",
" 809 | \n",
"
\n",
" \n",
" 3 | \n",
" 2019-01-18 | \n",
" 11860 | \n",
" 13056 | \n",
" NaN | \n",
" 1972 | \n",
" NaN | \n",
" 59 | \n",
" 29 | \n",
" 319 | \n",
" 1373 | \n",
" ... | \n",
" 627 | \n",
" 1566 | \n",
" 1197 | \n",
" 979 | \n",
" 1187 | \n",
" 1344 | \n",
" 1073 | \n",
" 1936 | \n",
" 1236 | \n",
" 683 | \n",
"
\n",
" \n",
" 4 | \n",
" 2019-01-25 | \n",
" 11740 | \n",
" 12486 | \n",
" NaN | \n",
" 1942 | \n",
" NaN | \n",
" 42 | \n",
" 22 | \n",
" 339 | \n",
" 1438 | \n",
" ... | \n",
" 640 | \n",
" 1593 | \n",
" 1137 | \n",
" 1037 | \n",
" 1131 | \n",
" 1291 | \n",
" 1045 | \n",
" 1875 | \n",
" 1229 | \n",
" 734 | \n",
"
\n",
" \n",
" 5 | \n",
" 2019-02-01 | \n",
" 11297 | \n",
" 11998 | \n",
" NaN | \n",
" 1931 | \n",
" NaN | \n",
" 57 | \n",
" 15 | \n",
" 307 | \n",
" 1367 | \n",
" ... | \n",
" 618 | \n",
" 1485 | \n",
" 1131 | \n",
" 963 | \n",
" 1150 | \n",
" 1202 | \n",
" 1066 | \n",
" 1718 | \n",
" 1191 | \n",
" 745 | \n",
"
\n",
" \n",
"
\n",
"
5 rows × 40 columns
\n",
"
"
],
"text/plain": [
"Week number Week ended Total deaths, all ages \\\n",
" Unnamed: 1_level_1 Unnamed: 2_level_1 \n",
"1 2019-01-04 10955 \n",
"2 2019-01-11 12609 \n",
"3 2019-01-18 11860 \n",
"4 2019-01-25 11740 \n",
"5 2019-02-01 11297 \n",
"\n",
"Week number Total deaths: average of corresponding Unnamed: 4_level_0 \\\n",
" Unnamed: 3_level_1 Deaths by underlying cause \n",
"1 12273 NaN \n",
"2 13670 NaN \n",
"3 13056 NaN \n",
"4 12486 NaN \n",
"5 11998 NaN \n",
"\n",
"Week number Unnamed: 5_level_0 \\\n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) \n",
"1 1736 \n",
"2 2214 \n",
"3 1972 \n",
"4 1942 \n",
"5 1931 \n",
"\n",
"Week number Persons Unnamed: 7_level_0 Unnamed: 8_level_0 \\\n",
" Deaths by age group Under 1 year 01-14 \n",
"1 NaN 43 15 \n",
"2 NaN 50 20 \n",
"3 NaN 59 29 \n",
"4 NaN 42 22 \n",
"5 NaN 57 15 \n",
"\n",
"Week number Unnamed: 9_level_0 Unnamed: 10_level_0 ... E12000001 E12000002 \\\n",
" 15-44 45-64 ... North East North West \n",
"1 215 1199 ... 574 1456 \n",
"2 280 1419 ... 643 1612 \n",
"3 319 1373 ... 627 1566 \n",
"4 339 1438 ... 640 1593 \n",
"5 307 1367 ... 618 1485 \n",
"\n",
"Week number E12000003 E12000004 E12000005 E12000006 \\\n",
" Yorkshire and The Humber East Midlands West Midlands East \n",
"1 1148 964 1180 1136 \n",
"2 1279 1089 1299 1386 \n",
"3 1197 979 1187 1344 \n",
"4 1137 1037 1131 1291 \n",
"5 1131 963 1150 1202 \n",
"\n",
"Week number E12000007 E12000008 E12000009 W92000004 \n",
" London South East South West Wales \n",
"1 1021 1586 1136 718 \n",
"2 1137 1927 1394 809 \n",
"3 1073 1936 1236 683 \n",
"4 1045 1875 1229 734 \n",
"5 1066 1718 1191 745 \n",
"\n",
"[5 rows x 40 columns]"
]
},
"execution_count": 78,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"raw_data_2019 = pd.read_csv('uk-deaths-data/publishedweek522019.csv', \n",
" parse_dates=[1], dayfirst=True,\n",
" index_col=0,\n",
" header=[0, 1])\n",
"raw_data_2019.head()"
]
},
{
"cell_type": "code",
"execution_count": 79,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" Week number | \n",
" Week ended | \n",
" Total deaths, all ages | \n",
" Total deaths: average of corresponding | \n",
" Unnamed: 4_level_0 | \n",
" Unnamed: 5_level_0 | \n",
" Persons | \n",
" Unnamed: 7_level_0 | \n",
" Unnamed: 8_level_0 | \n",
" Unnamed: 9_level_0 | \n",
" Unnamed: 10_level_0 | \n",
" ... | \n",
" E12000001 | \n",
" E12000002 | \n",
" E12000003 | \n",
" E12000004 | \n",
" E12000005 | \n",
" E12000006 | \n",
" E12000007 | \n",
" E12000008 | \n",
" E12000009 | \n",
" W92000004 | \n",
"
\n",
" \n",
" | \n",
" Unnamed: 1_level_1 | \n",
" Unnamed: 2_level_1 | \n",
" Unnamed: 3_level_1 | \n",
" Deaths by underlying cause | \n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) | \n",
" Deaths by age group | \n",
" Under 1 year | \n",
" 01-14 | \n",
" 15-44 | \n",
" 45-64 | \n",
" ... | \n",
" North East | \n",
" North West | \n",
" Yorkshire and The Humber | \n",
" East Midlands | \n",
" West Midlands | \n",
" East | \n",
" London | \n",
" South East | \n",
" South West | \n",
" Wales | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 2018-01-05 | \n",
" 12723 | \n",
" 12079 | \n",
" NaN | \n",
" 2361 | \n",
" NaN | \n",
" 52 | \n",
" 18 | \n",
" 208 | \n",
" 1290 | \n",
" ... | \n",
" 640 | \n",
" 1747 | \n",
" 1300 | \n",
" 1085 | \n",
" 1313 | \n",
" 1404 | \n",
" 1130 | \n",
" 1902 | \n",
" 1393 | \n",
" 783 | \n",
"
\n",
" \n",
" 2 | \n",
" 2018-01-12 | \n",
" 15050 | \n",
" 13167 | \n",
" NaN | \n",
" 3075 | \n",
" NaN | \n",
" 73 | \n",
" 17 | \n",
" 302 | \n",
" 1561 | \n",
" ... | \n",
" 837 | \n",
" 2113 | \n",
" 1481 | \n",
" 1266 | \n",
" 1468 | \n",
" 1677 | \n",
" 1342 | \n",
" 2371 | \n",
" 1552 | \n",
" 904 | \n",
"
\n",
" \n",
" 3 | \n",
" 2018-01-19 | \n",
" 14256 | \n",
" 12418 | \n",
" NaN | \n",
" 2829 | \n",
" NaN | \n",
" 59 | \n",
" 22 | \n",
" 286 | \n",
" 1507 | \n",
" ... | \n",
" 783 | \n",
" 1828 | \n",
" 1387 | \n",
" 1175 | \n",
" 1544 | \n",
" 1562 | \n",
" 1239 | \n",
" 2287 | \n",
" 1535 | \n",
" 885 | \n",
"
\n",
" \n",
" 4 | \n",
" 2018-01-26 | \n",
" 13935 | \n",
" 11954 | \n",
" NaN | \n",
" 2672 | \n",
" NaN | \n",
" 50 | \n",
" 25 | \n",
" 298 | \n",
" 1459 | \n",
" ... | \n",
" 779 | \n",
" 1855 | \n",
" 1363 | \n",
" 1165 | \n",
" 1433 | \n",
" 1553 | \n",
" 1264 | \n",
" 2184 | \n",
" 1460 | \n",
" 850 | \n",
"
\n",
" \n",
" 5 | \n",
" 2018-02-02 | \n",
" 13285 | \n",
" 11605 | \n",
" NaN | \n",
" 2479 | \n",
" NaN | \n",
" 41 | \n",
" 14 | \n",
" 339 | \n",
" 1404 | \n",
" ... | \n",
" 740 | \n",
" 1728 | \n",
" 1285 | \n",
" 1154 | \n",
" 1441 | \n",
" 1468 | \n",
" 1208 | \n",
" 2046 | \n",
" 1382 | \n",
" 815 | \n",
"
\n",
" \n",
"
\n",
"
5 rows × 40 columns
\n",
"
"
],
"text/plain": [
"Week number Week ended Total deaths, all ages \\\n",
" Unnamed: 1_level_1 Unnamed: 2_level_1 \n",
"1 2018-01-05 12723 \n",
"2 2018-01-12 15050 \n",
"3 2018-01-19 14256 \n",
"4 2018-01-26 13935 \n",
"5 2018-02-02 13285 \n",
"\n",
"Week number Total deaths: average of corresponding Unnamed: 4_level_0 \\\n",
" Unnamed: 3_level_1 Deaths by underlying cause \n",
"1 12079 NaN \n",
"2 13167 NaN \n",
"3 12418 NaN \n",
"4 11954 NaN \n",
"5 11605 NaN \n",
"\n",
"Week number Unnamed: 5_level_0 \\\n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) \n",
"1 2361 \n",
"2 3075 \n",
"3 2829 \n",
"4 2672 \n",
"5 2479 \n",
"\n",
"Week number Persons Unnamed: 7_level_0 Unnamed: 8_level_0 \\\n",
" Deaths by age group Under 1 year 01-14 \n",
"1 NaN 52 18 \n",
"2 NaN 73 17 \n",
"3 NaN 59 22 \n",
"4 NaN 50 25 \n",
"5 NaN 41 14 \n",
"\n",
"Week number Unnamed: 9_level_0 Unnamed: 10_level_0 ... E12000001 E12000002 \\\n",
" 15-44 45-64 ... North East North West \n",
"1 208 1290 ... 640 1747 \n",
"2 302 1561 ... 837 2113 \n",
"3 286 1507 ... 783 1828 \n",
"4 298 1459 ... 779 1855 \n",
"5 339 1404 ... 740 1728 \n",
"\n",
"Week number E12000003 E12000004 E12000005 E12000006 \\\n",
" Yorkshire and The Humber East Midlands West Midlands East \n",
"1 1300 1085 1313 1404 \n",
"2 1481 1266 1468 1677 \n",
"3 1387 1175 1544 1562 \n",
"4 1363 1165 1433 1553 \n",
"5 1285 1154 1441 1468 \n",
"\n",
"Week number E12000007 E12000008 E12000009 W92000004 \n",
" London South East South West Wales \n",
"1 1130 1902 1393 783 \n",
"2 1342 2371 1552 904 \n",
"3 1239 2287 1535 885 \n",
"4 1264 2184 1460 850 \n",
"5 1208 2046 1382 815 \n",
"\n",
"[5 rows x 40 columns]"
]
},
"execution_count": 79,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"raw_data_2018 = pd.read_csv('uk-deaths-data/publishedweek522018.csv', \n",
" parse_dates=[1], dayfirst=True,\n",
" index_col=0,\n",
" header=[0, 1])\n",
"raw_data_2018.head()"
]
},
{
"cell_type": "code",
"execution_count": 89,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" Week number | \n",
" Week ended | \n",
" Total deaths, all ages | \n",
" Total deaths: average of corresponding | \n",
" Unnamed: 4_level_0 | \n",
" Unnamed: 5_level_0 | \n",
" Persons | \n",
" Unnamed: 7_level_0 | \n",
" Unnamed: 8_level_0 | \n",
" Unnamed: 9_level_0 | \n",
" Unnamed: 10_level_0 | \n",
" ... | \n",
" E12000001 | \n",
" E12000002 | \n",
" E12000003 | \n",
" E12000004 | \n",
" E12000005 | \n",
" E12000006 | \n",
" E12000007 | \n",
" E12000008 | \n",
" E12000009 | \n",
" W92000004 | \n",
"
\n",
" \n",
" | \n",
" Unnamed: 1_level_1 | \n",
" Unnamed: 2_level_1 | \n",
" Unnamed: 3_level_1 | \n",
" Deaths by underlying cause | \n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) | \n",
" Deaths by age group | \n",
" Under 1 year | \n",
" 01-14 | \n",
" 15-44 | \n",
" 45-64 | \n",
" ... | \n",
" North East | \n",
" North West | \n",
" Yorkshire and The Humber | \n",
" East Midlands | \n",
" West Midlands | \n",
" East | \n",
" London | \n",
" South East | \n",
" South West | \n",
" Wales | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 2017-01-06 | \n",
" 11991 | \n",
" 11782 | \n",
" NaN | \n",
" 2321 | \n",
" NaN | \n",
" 58 | \n",
" 18 | \n",
" 215 | \n",
" 1232 | \n",
" ... | \n",
" 602 | \n",
" 1665 | \n",
" 1208 | \n",
" 1016 | \n",
" 1243 | \n",
" 1187 | \n",
" 1173 | \n",
" 1899 | \n",
" 1228 | \n",
" 744 | \n",
"
\n",
" \n",
" 2 | \n",
" 2017-01-13 | \n",
" 13715 | \n",
" 12692 | \n",
" NaN | \n",
" 2690 | \n",
" NaN | \n",
" 56 | \n",
" 18 | \n",
" 287 | \n",
" 1372 | \n",
" ... | \n",
" 738 | \n",
" 1840 | \n",
" 1351 | \n",
" 1170 | \n",
" 1439 | \n",
" 1450 | \n",
" 1296 | \n",
" 2161 | \n",
" 1419 | \n",
" 825 | \n",
"
\n",
" \n",
" 3 | \n",
" 2017-01-20 | \n",
" 13610 | \n",
" 11773 | \n",
" NaN | \n",
" 2625 | \n",
" NaN | \n",
" 54 | \n",
" 19 | \n",
" 301 | \n",
" 1374 | \n",
" ... | \n",
" 676 | \n",
" 1786 | \n",
" 1383 | \n",
" 1190 | \n",
" 1414 | \n",
" 1484 | \n",
" 1276 | \n",
" 2195 | \n",
" 1346 | \n",
" 835 | \n",
"
\n",
" \n",
" 4 | \n",
" 2017-01-27 | \n",
" 12877 | \n",
" 11441 | \n",
" NaN | \n",
" 2373 | \n",
" NaN | \n",
" 52 | \n",
" 16 | \n",
" 286 | \n",
" 1273 | \n",
" ... | \n",
" 627 | \n",
" 1666 | \n",
" 1242 | \n",
" 1116 | \n",
" 1362 | \n",
" 1488 | \n",
" 1198 | \n",
" 1999 | \n",
" 1278 | \n",
" 881 | \n",
"
\n",
" \n",
" 5 | \n",
" 2017-02-03 | \n",
" 12485 | \n",
" 11128 | \n",
" NaN | \n",
" 2147 | \n",
" NaN | \n",
" 55 | \n",
" 14 | \n",
" 308 | \n",
" 1239 | \n",
" ... | \n",
" 631 | \n",
" 1625 | \n",
" 1243 | \n",
" 1045 | \n",
" 1293 | \n",
" 1384 | \n",
" 1210 | \n",
" 1920 | \n",
" 1359 | \n",
" 749 | \n",
"
\n",
" \n",
"
\n",
"
5 rows × 40 columns
\n",
"
"
],
"text/plain": [
"Week number Week ended Total deaths, all ages \\\n",
" Unnamed: 1_level_1 Unnamed: 2_level_1 \n",
"1 2017-01-06 11991 \n",
"2 2017-01-13 13715 \n",
"3 2017-01-20 13610 \n",
"4 2017-01-27 12877 \n",
"5 2017-02-03 12485 \n",
"\n",
"Week number Total deaths: average of corresponding Unnamed: 4_level_0 \\\n",
" Unnamed: 3_level_1 Deaths by underlying cause \n",
"1 11782 NaN \n",
"2 12692 NaN \n",
"3 11773 NaN \n",
"4 11441 NaN \n",
"5 11128 NaN \n",
"\n",
"Week number Unnamed: 5_level_0 \\\n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) \n",
"1 2321 \n",
"2 2690 \n",
"3 2625 \n",
"4 2373 \n",
"5 2147 \n",
"\n",
"Week number Persons Unnamed: 7_level_0 Unnamed: 8_level_0 \\\n",
" Deaths by age group Under 1 year 01-14 \n",
"1 NaN 58 18 \n",
"2 NaN 56 18 \n",
"3 NaN 54 19 \n",
"4 NaN 52 16 \n",
"5 NaN 55 14 \n",
"\n",
"Week number Unnamed: 9_level_0 Unnamed: 10_level_0 ... E12000001 E12000002 \\\n",
" 15-44 45-64 ... North East North West \n",
"1 215 1232 ... 602 1665 \n",
"2 287 1372 ... 738 1840 \n",
"3 301 1374 ... 676 1786 \n",
"4 286 1273 ... 627 1666 \n",
"5 308 1239 ... 631 1625 \n",
"\n",
"Week number E12000003 E12000004 E12000005 E12000006 \\\n",
" Yorkshire and The Humber East Midlands West Midlands East \n",
"1 1208 1016 1243 1187 \n",
"2 1351 1170 1439 1450 \n",
"3 1383 1190 1414 1484 \n",
"4 1242 1116 1362 1488 \n",
"5 1243 1045 1293 1384 \n",
"\n",
"Week number E12000007 E12000008 E12000009 W92000004 \n",
" London South East South West Wales \n",
"1 1173 1899 1228 744 \n",
"2 1296 2161 1419 825 \n",
"3 1276 2195 1346 835 \n",
"4 1198 1999 1278 881 \n",
"5 1210 1920 1359 749 \n",
"\n",
"[5 rows x 40 columns]"
]
},
"execution_count": 89,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"raw_data_2017 = pd.read_csv('uk-deaths-data/publishedweek522017.csv', \n",
" parse_dates=[1], dayfirst=True,\n",
" index_col=0,\n",
" header=[0, 1])\n",
"raw_data_2017.head()"
]
},
{
"cell_type": "code",
"execution_count": 98,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" Week number | \n",
" Week ended | \n",
" Total deaths, all ages | \n",
" Total deaths: average of corresponding | \n",
" Unnamed: 4_level_0 | \n",
" Unnamed: 5_level_0 | \n",
" Persons | \n",
" Unnamed: 7_level_0 | \n",
" Unnamed: 8_level_0 | \n",
" Unnamed: 9_level_0 | \n",
" Unnamed: 10_level_0 | \n",
" ... | \n",
" E12000001 | \n",
" E12000002 | \n",
" E12000003 | \n",
" E12000004 | \n",
" E12000005 | \n",
" E12000006 | \n",
" E12000007 | \n",
" E12000008 | \n",
" E12000009 | \n",
" W92000004 | \n",
"
\n",
" \n",
" | \n",
" Unnamed: 1_level_1 | \n",
" Unnamed: 2_level_1 | \n",
" Unnamed: 3_level_1 | \n",
" Deaths by underlying cause | \n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) | \n",
" Deaths by age group | \n",
" Under 1 year | \n",
" 01-14 | \n",
" 15-44 | \n",
" 45-64 | \n",
" ... | \n",
" North East | \n",
" North West | \n",
" Yorkshire and The Humber | \n",
" East Midlands | \n",
" West Midlands | \n",
" East | \n",
" London | \n",
" South East | \n",
" South West | \n",
" Wales | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 2016-01-08 | \n",
" 13045 | \n",
" 11701.4 | \n",
" NaN | \n",
" 2046 | \n",
" NaN | \n",
" 49 | \n",
" 17 | \n",
" 299 | \n",
" 1482 | \n",
" ... | \n",
" 705 | \n",
" 1748 | \n",
" 1284 | \n",
" 1067 | \n",
" 1396 | \n",
" 1401 | \n",
" 1226 | \n",
" 1951 | \n",
" 1424 | \n",
" 809 | \n",
"
\n",
" \n",
" 2 | \n",
" 2016-01-15 | \n",
" 11501 | \n",
" 13016.4 | \n",
" NaN | \n",
" 1835 | \n",
" NaN | \n",
" 64 | \n",
" 24 | \n",
" 326 | \n",
" 1363 | \n",
" ... | \n",
" 600 | \n",
" 1532 | \n",
" 1148 | \n",
" 956 | \n",
" 1208 | \n",
" 1237 | \n",
" 1048 | \n",
" 1771 | \n",
" 1263 | \n",
" 711 | \n",
"
\n",
" \n",
" 3 | \n",
" 2016-01-22 | \n",
" 11473 | \n",
" 11765.4 | \n",
" NaN | \n",
" 1775 | \n",
" NaN | \n",
" 50 | \n",
" 18 | \n",
" 345 | \n",
" 1293 | \n",
" ... | \n",
" 612 | \n",
" 1602 | \n",
" 1119 | \n",
" 929 | \n",
" 1209 | \n",
" 1253 | \n",
" 1068 | \n",
" 1710 | \n",
" 1219 | \n",
" 720 | \n",
"
\n",
" \n",
" 4 | \n",
" 2016-01-29 | \n",
" 11317 | \n",
" 11289.0 | \n",
" NaN | \n",
" 1810 | \n",
" NaN | \n",
" 47 | \n",
" 14 | \n",
" 324 | \n",
" 1322 | \n",
" ... | \n",
" 631 | \n",
" 1516 | \n",
" 1131 | \n",
" 858 | \n",
" 1195 | \n",
" 1236 | \n",
" 1065 | \n",
" 1730 | \n",
" 1207 | \n",
" 717 | \n",
"
\n",
" \n",
" 5 | \n",
" 2016-02-05 | \n",
" 11052 | \n",
" 10965.6 | \n",
" NaN | \n",
" 1748 | \n",
" NaN | \n",
" 60 | \n",
" 22 | \n",
" 314 | \n",
" 1281 | \n",
" ... | \n",
" 620 | \n",
" 1459 | \n",
" 1109 | \n",
" 954 | \n",
" 1197 | \n",
" 1160 | \n",
" 1059 | \n",
" 1604 | \n",
" 1169 | \n",
" 690 | \n",
"
\n",
" \n",
"
\n",
"
5 rows × 40 columns
\n",
"
"
],
"text/plain": [
"Week number Week ended Total deaths, all ages \\\n",
" Unnamed: 1_level_1 Unnamed: 2_level_1 \n",
"1 2016-01-08 13045 \n",
"2 2016-01-15 11501 \n",
"3 2016-01-22 11473 \n",
"4 2016-01-29 11317 \n",
"5 2016-02-05 11052 \n",
"\n",
"Week number Total deaths: average of corresponding Unnamed: 4_level_0 \\\n",
" Unnamed: 3_level_1 Deaths by underlying cause \n",
"1 11701.4 NaN \n",
"2 13016.4 NaN \n",
"3 11765.4 NaN \n",
"4 11289.0 NaN \n",
"5 10965.6 NaN \n",
"\n",
"Week number Unnamed: 5_level_0 \\\n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) \n",
"1 2046 \n",
"2 1835 \n",
"3 1775 \n",
"4 1810 \n",
"5 1748 \n",
"\n",
"Week number Persons Unnamed: 7_level_0 Unnamed: 8_level_0 \\\n",
" Deaths by age group Under 1 year 01-14 \n",
"1 NaN 49 17 \n",
"2 NaN 64 24 \n",
"3 NaN 50 18 \n",
"4 NaN 47 14 \n",
"5 NaN 60 22 \n",
"\n",
"Week number Unnamed: 9_level_0 Unnamed: 10_level_0 ... E12000001 E12000002 \\\n",
" 15-44 45-64 ... North East North West \n",
"1 299 1482 ... 705 1748 \n",
"2 326 1363 ... 600 1532 \n",
"3 345 1293 ... 612 1602 \n",
"4 324 1322 ... 631 1516 \n",
"5 314 1281 ... 620 1459 \n",
"\n",
"Week number E12000003 E12000004 E12000005 E12000006 \\\n",
" Yorkshire and The Humber East Midlands West Midlands East \n",
"1 1284 1067 1396 1401 \n",
"2 1148 956 1208 1237 \n",
"3 1119 929 1209 1253 \n",
"4 1131 858 1195 1236 \n",
"5 1109 954 1197 1160 \n",
"\n",
"Week number E12000007 E12000008 E12000009 W92000004 \n",
" London South East South West Wales \n",
"1 1226 1951 1424 809 \n",
"2 1048 1771 1263 711 \n",
"3 1068 1710 1219 720 \n",
"4 1065 1730 1207 717 \n",
"5 1059 1604 1169 690 \n",
"\n",
"[5 rows x 40 columns]"
]
},
"execution_count": 98,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"raw_data_2016 = pd.read_csv('uk-deaths-data/publishedweek522016.csv', \n",
" parse_dates=[1], dayfirst=True,\n",
" index_col=0,\n",
" header=[0, 1])\n",
"raw_data_2016.head()"
]
},
{
"cell_type": "code",
"execution_count": 114,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" Week number | \n",
" Week ended | \n",
" Total deaths, all ages | \n",
" Total deaths: average of corresponding | \n",
" Unnamed: 4_level_0 | \n",
" Unnamed: 5_level_0 | \n",
" Persons | \n",
" Unnamed: 7_level_0 | \n",
" Unnamed: 8_level_0 | \n",
" Unnamed: 9_level_0 | \n",
" Unnamed: 10_level_0 | \n",
" ... | \n",
" E12000001 | \n",
" E12000002 | \n",
" E12000003 | \n",
" E12000004 | \n",
" E12000005 | \n",
" E12000006 | \n",
" E12000007 | \n",
" E12000008 | \n",
" E12000009 | \n",
" W92000004 | \n",
"
\n",
" \n",
" | \n",
" Unnamed: 1_level_1 | \n",
" Unnamed: 2_level_1 | \n",
" Unnamed: 3_level_1 | \n",
" Deaths by underlying cause | \n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) | \n",
" Deaths by age group | \n",
" Under 1 year | \n",
" 01-14 | \n",
" 15-44 | \n",
" 45-64 | \n",
" ... | \n",
" North East | \n",
" North West | \n",
" Yorkshire and The Humber | \n",
" East Midlands | \n",
" West Midlands | \n",
" East | \n",
" London | \n",
" South East | \n",
" South West | \n",
" Wales | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 2015-01-02 | \n",
" 12286 | \n",
" 11837.6 | \n",
" NaN | \n",
" 2418 | \n",
" NaN | \n",
" 56 | \n",
" 16 | \n",
" 198 | \n",
" 1229 | \n",
" ... | \n",
" 699 | \n",
" 1718 | \n",
" 1261 | \n",
" 1012 | \n",
" 1296 | \n",
" 1221 | \n",
" 1233 | \n",
" 1811 | \n",
" 1272 | \n",
" 725 | \n",
"
\n",
" \n",
" 2 | \n",
" 2015-01-09 | \n",
" 16237 | \n",
" 12277.0 | \n",
" NaN | \n",
" 3521 | \n",
" NaN | \n",
" 80 | \n",
" 28 | \n",
" 298 | \n",
" 1552 | \n",
" ... | \n",
" 753 | \n",
" 2282 | \n",
" 1607 | \n",
" 1328 | \n",
" 1713 | \n",
" 1712 | \n",
" 1549 | \n",
" 2525 | \n",
" 1697 | \n",
" 1031 | \n",
"
\n",
" \n",
" 3 | \n",
" 2015-01-16 | \n",
" 14866 | \n",
" 11145.0 | \n",
" NaN | \n",
" 3251 | \n",
" NaN | \n",
" 46 | \n",
" 28 | \n",
" 316 | \n",
" 1393 | \n",
" ... | \n",
" 799 | \n",
" 1968 | \n",
" 1442 | \n",
" 1286 | \n",
" 1542 | \n",
" 1657 | \n",
" 1275 | \n",
" 2427 | \n",
" 1498 | \n",
" 936 | \n",
"
\n",
" \n",
" 4 | \n",
" 2015-01-23 | \n",
" 13934 | \n",
" 10714.0 | \n",
" NaN | \n",
" 2734 | \n",
" NaN | \n",
" 61 | \n",
" 20 | \n",
" 318 | \n",
" 1387 | \n",
" ... | \n",
" 798 | \n",
" 1806 | \n",
" 1386 | \n",
" 1206 | \n",
" 1400 | \n",
" 1565 | \n",
" 1220 | \n",
" 2196 | \n",
" 1510 | \n",
" 828 | \n",
"
\n",
" \n",
" 5 | \n",
" 2015-01-30 | \n",
" 12900 | \n",
" 10492.0 | \n",
" NaN | \n",
" 2278 | \n",
" NaN | \n",
" 66 | \n",
" 20 | \n",
" 329 | \n",
" 1368 | \n",
" ... | \n",
" 717 | \n",
" 1669 | \n",
" 1220 | \n",
" 1119 | \n",
" 1395 | \n",
" 1450 | \n",
" 1119 | \n",
" 1991 | \n",
" 1398 | \n",
" 801 | \n",
"
\n",
" \n",
"
\n",
"
5 rows × 40 columns
\n",
"
"
],
"text/plain": [
"Week number Week ended Total deaths, all ages \\\n",
" Unnamed: 1_level_1 Unnamed: 2_level_1 \n",
"1 2015-01-02 12286 \n",
"2 2015-01-09 16237 \n",
"3 2015-01-16 14866 \n",
"4 2015-01-23 13934 \n",
"5 2015-01-30 12900 \n",
"\n",
"Week number Total deaths: average of corresponding Unnamed: 4_level_0 \\\n",
" Unnamed: 3_level_1 Deaths by underlying cause \n",
"1 11837.6 NaN \n",
"2 12277.0 NaN \n",
"3 11145.0 NaN \n",
"4 10714.0 NaN \n",
"5 10492.0 NaN \n",
"\n",
"Week number Unnamed: 5_level_0 \\\n",
" All respiratory diseases (ICD-10 J00-J99)\\nICD-10 v 2013 (IRIS) \n",
"1 2418 \n",
"2 3521 \n",
"3 3251 \n",
"4 2734 \n",
"5 2278 \n",
"\n",
"Week number Persons Unnamed: 7_level_0 Unnamed: 8_level_0 \\\n",
" Deaths by age group Under 1 year 01-14 \n",
"1 NaN 56 16 \n",
"2 NaN 80 28 \n",
"3 NaN 46 28 \n",
"4 NaN 61 20 \n",
"5 NaN 66 20 \n",
"\n",
"Week number Unnamed: 9_level_0 Unnamed: 10_level_0 ... E12000001 E12000002 \\\n",
" 15-44 45-64 ... North East North West \n",
"1 198 1229 ... 699 1718 \n",
"2 298 1552 ... 753 2282 \n",
"3 316 1393 ... 799 1968 \n",
"4 318 1387 ... 798 1806 \n",
"5 329 1368 ... 717 1669 \n",
"\n",
"Week number E12000003 E12000004 E12000005 E12000006 \\\n",
" Yorkshire and The Humber East Midlands West Midlands East \n",
"1 1261 1012 1296 1221 \n",
"2 1607 1328 1713 1712 \n",
"3 1442 1286 1542 1657 \n",
"4 1386 1206 1400 1565 \n",
"5 1220 1119 1395 1450 \n",
"\n",
"Week number E12000007 E12000008 E12000009 W92000004 \n",
" London South East South West Wales \n",
"1 1233 1811 1272 725 \n",
"2 1549 2525 1697 1031 \n",
"3 1275 2427 1498 936 \n",
"4 1220 2196 1510 828 \n",
"5 1119 1991 1398 801 \n",
"\n",
"[5 rows x 40 columns]"
]
},
"execution_count": 114,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"raw_data_2015 = pd.read_csv('uk-deaths-data/publishedweek2015.csv', \n",
" parse_dates=[1], dayfirst=True,\n",
" index_col=0,\n",
" header=[0, 1])\n",
"raw_data_2015.head()"
]
},
{
"cell_type": "code",
"execution_count": 115,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" total_2020 | \n",
" total_previous | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 12254 | \n",
" 12175 | \n",
"
\n",
" \n",
" 2 | \n",
" 14058 | \n",
" 13822 | \n",
"
\n",
" \n",
" 3 | \n",
" 12990 | \n",
" 13216 | \n",
"
\n",
" \n",
" 4 | \n",
" 11856 | \n",
" 12760 | \n",
"
\n",
" \n",
" 5 | \n",
" 11612 | \n",
" 12206 | \n",
"
\n",
" \n",
" 6 | \n",
" 10986 | \n",
" 11925 | \n",
"
\n",
" \n",
" 7 | \n",
" 10944 | \n",
" 11627 | \n",
"
\n",
" \n",
" 8 | \n",
" 10841 | \n",
" 11548 | \n",
"
\n",
" \n",
" 9 | \n",
" 10816 | \n",
" 11183 | \n",
"
\n",
" \n",
" 10 | \n",
" 10895 | \n",
" 11498 | \n",
"
\n",
" \n",
" 11 | \n",
" 11019 | \n",
" 11205 | \n",
"
\n",
" \n",
" 12 | \n",
" 10645 | \n",
" 10573 | \n",
"
\n",
" \n",
" 13 | \n",
" 11141 | \n",
" 10130 | \n",
"
\n",
" \n",
" 14 | \n",
" 16387 | \n",
" 10305 | \n",
"
\n",
" \n",
" 15 | \n",
" 18516 | \n",
" 10520 | \n",
"
\n",
" \n",
" 16 | \n",
" 22351 | \n",
" 10497 | \n",
"
\n",
" \n",
" 17 | \n",
" 21997 | \n",
" 10458 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" total_2020 total_previous\n",
"1 12254 12175\n",
"2 14058 13822\n",
"3 12990 13216\n",
"4 11856 12760\n",
"5 11612 12206\n",
"6 10986 11925\n",
"7 10944 11627\n",
"8 10841 11548\n",
"9 10816 11183\n",
"10 10895 11498\n",
"11 11019 11205\n",
"12 10645 10573\n",
"13 11141 10130\n",
"14 16387 10305\n",
"15 18516 10520\n",
"16 22351 10497\n",
"17 21997 10458"
]
},
"execution_count": 115,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"deaths_headlines = raw_data_2020.iloc[:, [1, 2]]\n",
"deaths_headlines.columns = ['total_2020', 'total_previous']\n",
"deaths_headlines"
]
},
{
"cell_type": "code",
"execution_count": 116,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" total_2019 | \n",
" total_previous | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 10955 | \n",
" 12273 | \n",
"
\n",
" \n",
" 2 | \n",
" 12609 | \n",
" 13670 | \n",
"
\n",
" \n",
" 3 | \n",
" 11860 | \n",
" 13056 | \n",
"
\n",
" \n",
" 4 | \n",
" 11740 | \n",
" 12486 | \n",
"
\n",
" \n",
" 5 | \n",
" 11297 | \n",
" 11998 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" total_2019 total_previous\n",
"1 10955 12273\n",
"2 12609 13670\n",
"3 11860 13056\n",
"4 11740 12486\n",
"5 11297 11998"
]
},
"execution_count": 116,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dh19 = raw_data_2019.iloc[:, [1, 2]]\n",
"dh19.columns = ['total_2019', 'total_previous']\n",
"dh19.head()"
]
},
{
"cell_type": "code",
"execution_count": 117,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" total_2018 | \n",
" total_previous | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 12723 | \n",
" 12079 | \n",
"
\n",
" \n",
" 2 | \n",
" 15050 | \n",
" 13167 | \n",
"
\n",
" \n",
" 3 | \n",
" 14256 | \n",
" 12418 | \n",
"
\n",
" \n",
" 4 | \n",
" 13935 | \n",
" 11954 | \n",
"
\n",
" \n",
" 5 | \n",
" 13285 | \n",
" 11605 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" total_2018 total_previous\n",
"1 12723 12079\n",
"2 15050 13167\n",
"3 14256 12418\n",
"4 13935 11954\n",
"5 13285 11605"
]
},
"execution_count": 117,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dh18 = raw_data_2018.iloc[:, [1, 2]]\n",
"dh18.columns = ['total_2018', 'total_previous']\n",
"dh18.head()"
]
},
{
"cell_type": "code",
"execution_count": 118,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" total_2017 | \n",
" total_previous | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 11991 | \n",
" 11782 | \n",
"
\n",
" \n",
" 2 | \n",
" 13715 | \n",
" 12692 | \n",
"
\n",
" \n",
" 3 | \n",
" 13610 | \n",
" 11773 | \n",
"
\n",
" \n",
" 4 | \n",
" 12877 | \n",
" 11441 | \n",
"
\n",
" \n",
" 5 | \n",
" 12485 | \n",
" 11128 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" total_2017 total_previous\n",
"1 11991 11782\n",
"2 13715 12692\n",
"3 13610 11773\n",
"4 12877 11441\n",
"5 12485 11128"
]
},
"execution_count": 118,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dh17 = raw_data_2017.iloc[:, [1, 2]]\n",
"dh17.columns = ['total_2017', 'total_previous']\n",
"dh17.head()"
]
},
{
"cell_type": "code",
"execution_count": 119,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" total_2016 | \n",
" total_previous | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 13045 | \n",
" 11701.4 | \n",
"
\n",
" \n",
" 2 | \n",
" 11501 | \n",
" 13016.4 | \n",
"
\n",
" \n",
" 3 | \n",
" 11473 | \n",
" 11765.4 | \n",
"
\n",
" \n",
" 4 | \n",
" 11317 | \n",
" 11289.0 | \n",
"
\n",
" \n",
" 5 | \n",
" 11052 | \n",
" 10965.6 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" total_2016 total_previous\n",
"1 13045 11701.4\n",
"2 11501 13016.4\n",
"3 11473 11765.4\n",
"4 11317 11289.0\n",
"5 11052 10965.6"
]
},
"execution_count": 119,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dh16 = raw_data_2016.iloc[:, [1, 2]]\n",
"dh16.columns = ['total_2016', 'total_previous']\n",
"dh16.head()"
]
},
{
"cell_type": "code",
"execution_count": 120,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" total_2015 | \n",
" total_previous | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 12286 | \n",
" 11837.6 | \n",
"
\n",
" \n",
" 2 | \n",
" 16237 | \n",
" 12277.0 | \n",
"
\n",
" \n",
" 3 | \n",
" 14866 | \n",
" 11145.0 | \n",
"
\n",
" \n",
" 4 | \n",
" 13934 | \n",
" 10714.0 | \n",
"
\n",
" \n",
" 5 | \n",
" 12900 | \n",
" 10492.0 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" total_2015 total_previous\n",
"1 12286 11837.6\n",
"2 16237 12277.0\n",
"3 14866 11145.0\n",
"4 13934 10714.0\n",
"5 12900 10492.0"
]
},
"execution_count": 120,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dh15 = raw_data_2015.iloc[:, [1, 2]]\n",
"dh15.columns = ['total_2015', 'total_previous']\n",
"dh15.head()"
]
},
{
"cell_type": "code",
"execution_count": 121,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" total_2020 | \n",
" total_previous | \n",
" total_2019 | \n",
" total_2018 | \n",
" total_2017 | \n",
" total_2016 | \n",
" total_2015 | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 12254.0 | \n",
" 12175.0 | \n",
" 10955.0 | \n",
" 12723.0 | \n",
" 11991.0 | \n",
" 13045.0 | \n",
" 12286 | \n",
"
\n",
" \n",
" 2 | \n",
" 14058.0 | \n",
" 13822.0 | \n",
" 12609.0 | \n",
" 15050.0 | \n",
" 13715.0 | \n",
" 11501.0 | \n",
" 16237 | \n",
"
\n",
" \n",
" 3 | \n",
" 12990.0 | \n",
" 13216.0 | \n",
" 11860.0 | \n",
" 14256.0 | \n",
" 13610.0 | \n",
" 11473.0 | \n",
" 14866 | \n",
"
\n",
" \n",
" 4 | \n",
" 11856.0 | \n",
" 12760.0 | \n",
" 11740.0 | \n",
" 13935.0 | \n",
" 12877.0 | \n",
" 11317.0 | \n",
" 13934 | \n",
"
\n",
" \n",
" 5 | \n",
" 11612.0 | \n",
" 12206.0 | \n",
" 11297.0 | \n",
" 13285.0 | \n",
" 12485.0 | \n",
" 11052.0 | \n",
" 12900 | \n",
"
\n",
" \n",
" 6 | \n",
" 10986.0 | \n",
" 11925.0 | \n",
" 11660.0 | \n",
" 12495.0 | \n",
" 12269.0 | \n",
" 11170.0 | \n",
" 12039 | \n",
"
\n",
" \n",
" 7 | \n",
" 10944.0 | \n",
" 11627.0 | \n",
" 11824.0 | \n",
" 12246.0 | \n",
" 11644.0 | \n",
" 10590.0 | \n",
" 11822 | \n",
"
\n",
" \n",
" 8 | \n",
" 10841.0 | \n",
" 11548.0 | \n",
" 11295.0 | \n",
" 12142.0 | \n",
" 11794.0 | \n",
" 11056.0 | \n",
" 11434 | \n",
"
\n",
" \n",
" 9 | \n",
" 10816.0 | \n",
" 11183.0 | \n",
" 11044.0 | \n",
" 10854.0 | \n",
" 11248.0 | \n",
" 11285.0 | \n",
" 11472 | \n",
"
\n",
" \n",
" 10 | \n",
" 10895.0 | \n",
" 11498.0 | \n",
" 10898.0 | \n",
" 12997.0 | \n",
" 11077.0 | \n",
" 11010.0 | \n",
" 11469 | \n",
"
\n",
" \n",
" 11 | \n",
" 11019.0 | \n",
" 11205.0 | \n",
" 10567.0 | \n",
" 12788.0 | \n",
" 10697.0 | \n",
" 11022.0 | \n",
" 10951 | \n",
"
\n",
" \n",
" 12 | \n",
" 10645.0 | \n",
" 10573.0 | \n",
" 10402.0 | \n",
" 11913.0 | \n",
" 10325.0 | \n",
" 9635.0 | \n",
" 10568 | \n",
"
\n",
" \n",
" 13 | \n",
" 11141.0 | \n",
" 10130.0 | \n",
" 9867.0 | \n",
" 9941.0 | \n",
" 10027.0 | \n",
" 10286.0 | \n",
" 10493 | \n",
"
\n",
" \n",
" 14 | \n",
" 16387.0 | \n",
" 10305.0 | \n",
" 10126.0 | \n",
" 10794.0 | \n",
" 9939.0 | \n",
" 11599.0 | \n",
" 9062 | \n",
"
\n",
" \n",
" 15 | \n",
" 18516.0 | \n",
" 10520.0 | \n",
" 10291.0 | \n",
" 12301.0 | \n",
" 8493.0 | \n",
" 11417.0 | \n",
" 10089 | \n",
"
\n",
" \n",
" 16 | \n",
" 22351.0 | \n",
" 10497.0 | \n",
" 9025.0 | \n",
" 11223.0 | \n",
" 9644.0 | \n",
" 10925.0 | \n",
" 11639 | \n",
"
\n",
" \n",
" 17 | \n",
" 21997.0 | \n",
" 10458.0 | \n",
" 10059.0 | \n",
" 10306.0 | \n",
" 10908.0 | \n",
" 10413.0 | \n",
" 10599 | \n",
"
\n",
" \n",
" 18 | \n",
" NaN | \n",
" NaN | \n",
" 11207.0 | \n",
" 10153.0 | \n",
" 9064.0 | \n",
" 9137.0 | \n",
" 10134 | \n",
"
\n",
" \n",
" 19 | \n",
" NaN | \n",
" NaN | \n",
" 9055.0 | \n",
" 8624.0 | \n",
" 10693.0 | \n",
" 10637.0 | \n",
" 8862 | \n",
"
\n",
" \n",
" 20 | \n",
" NaN | \n",
" NaN | \n",
" 10272.0 | \n",
" 10141.0 | \n",
" 10288.0 | \n",
" 9953.0 | \n",
" 10290 | \n",
"
\n",
" \n",
" 21 | \n",
" NaN | \n",
" NaN | \n",
" 10284.0 | \n",
" 9636.0 | \n",
" 10040.0 | \n",
" 9739.0 | \n",
" 10005 | \n",
"
\n",
" \n",
" 22 | \n",
" NaN | \n",
" NaN | \n",
" 8260.0 | \n",
" 8147.0 | \n",
" 8332.0 | \n",
" 7909.0 | \n",
" 8213 | \n",
"
\n",
" \n",
" 23 | \n",
" NaN | \n",
" NaN | \n",
" 10140.0 | \n",
" 9950.0 | \n",
" 9766.0 | \n",
" 9873.0 | \n",
" 10157 | \n",
"
\n",
" \n",
" 24 | \n",
" NaN | \n",
" NaN | \n",
" 9445.0 | \n",
" 9343.0 | \n",
" 9367.0 | \n",
" 9386.0 | \n",
" 9548 | \n",
"
\n",
" \n",
" 25 | \n",
" NaN | \n",
" NaN | \n",
" 9458.0 | \n",
" 9256.0 | \n",
" 9627.0 | \n",
" 9365.0 | \n",
" 9312 | \n",
"
\n",
" \n",
" 26 | \n",
" NaN | \n",
" NaN | \n",
" 9511.0 | \n",
" 9212.0 | \n",
" 9334.0 | \n",
" 9228.0 | \n",
" 9190 | \n",
"
\n",
" \n",
" 27 | \n",
" NaN | \n",
" NaN | \n",
" 9062.0 | \n",
" 9258.0 | \n",
" 9263.0 | \n",
" 9138.0 | \n",
" 9205 | \n",
"
\n",
" \n",
" 28 | \n",
" NaN | \n",
" NaN | \n",
" 9179.0 | \n",
" 9293.0 | \n",
" 9376.0 | \n",
" 9388.0 | \n",
" 9015 | \n",
"
\n",
" \n",
" 29 | \n",
" NaN | \n",
" NaN | \n",
" 9080.0 | \n",
" 9127.0 | \n",
" 9113.0 | \n",
" 9350.0 | \n",
" 8802 | \n",
"
\n",
" \n",
" 30 | \n",
" NaN | \n",
" NaN | \n",
" 9112.0 | \n",
" 9141.0 | \n",
" 8882.0 | \n",
" 9335.0 | \n",
" 8791 | \n",
"
\n",
" \n",
" 31 | \n",
" NaN | \n",
" NaN | \n",
" 9271.0 | \n",
" 9161.0 | \n",
" 8941.0 | \n",
" 9182.0 | \n",
" 8617 | \n",
"
\n",
" \n",
" 32 | \n",
" NaN | \n",
" NaN | \n",
" 9122.0 | \n",
" 9319.0 | \n",
" 9038.0 | \n",
" 9172.0 | \n",
" 8862 | \n",
"
\n",
" \n",
" 33 | \n",
" NaN | \n",
" NaN | \n",
" 9093.0 | \n",
" 8830.0 | \n",
" 9299.0 | \n",
" 9070.0 | \n",
" 9148 | \n",
"
\n",
" \n",
" 34 | \n",
" NaN | \n",
" NaN | \n",
" 8994.0 | \n",
" 8978.0 | \n",
" 9382.0 | \n",
" 9319.0 | \n",
" 9121 | \n",
"
\n",
" \n",
" 35 | \n",
" NaN | \n",
" NaN | \n",
" 8242.0 | \n",
" 7865.0 | \n",
" 8149.0 | \n",
" 7923.0 | \n",
" 9026 | \n",
"
\n",
" \n",
" 36 | \n",
" NaN | \n",
" NaN | \n",
" 9695.0 | \n",
" 9445.0 | \n",
" 9497.0 | \n",
" 9399.0 | \n",
" 7878 | \n",
"
\n",
" \n",
" 37 | \n",
" NaN | \n",
" NaN | \n",
" 9513.0 | \n",
" 9191.0 | \n",
" 9454.0 | \n",
" 9124.0 | \n",
" 9258 | \n",
"
\n",
" \n",
" 38 | \n",
" NaN | \n",
" NaN | \n",
" 9440.0 | \n",
" 9305.0 | \n",
" 9534.0 | \n",
" 8945.0 | \n",
" 9097 | \n",
"
\n",
" \n",
" 39 | \n",
" NaN | \n",
" NaN | \n",
" 9517.0 | \n",
" 9150.0 | \n",
" 9689.0 | \n",
" 8994.0 | \n",
" 9529 | \n",
"
\n",
" \n",
" 40 | \n",
" NaN | \n",
" NaN | \n",
" 9799.0 | \n",
" 9503.0 | \n",
" 9778.0 | \n",
" 9291.0 | \n",
" 9410 | \n",
"
\n",
" \n",
" 41 | \n",
" NaN | \n",
" NaN | \n",
" 9973.0 | \n",
" 9649.0 | \n",
" 9940.0 | \n",
" 9719.0 | \n",
" 9776 | \n",
"
\n",
" \n",
" 42 | \n",
" NaN | \n",
" NaN | \n",
" 10156.0 | \n",
" 9864.0 | \n",
" 10031.0 | \n",
" 9768.0 | \n",
" 9511 | \n",
"
\n",
" \n",
" 43 | \n",
" NaN | \n",
" NaN | \n",
" 10021.0 | \n",
" 9603.0 | \n",
" 9739.0 | \n",
" 9724.0 | \n",
" 9711 | \n",
"
\n",
" \n",
" 44 | \n",
" NaN | \n",
" NaN | \n",
" 10164.0 | \n",
" 9529.0 | \n",
" 9984.0 | \n",
" 10152.0 | \n",
" 9618 | \n",
"
\n",
" \n",
" 45 | \n",
" NaN | \n",
" NaN | \n",
" 10697.0 | \n",
" 10151.0 | \n",
" 10346.0 | \n",
" 10470.0 | \n",
" 9994 | \n",
"
\n",
" \n",
" 46 | \n",
" NaN | \n",
" NaN | \n",
" 10650.0 | \n",
" 10193.0 | \n",
" 10275.0 | \n",
" 10694.0 | \n",
" 9938 | \n",
"
\n",
" \n",
" 47 | \n",
" NaN | \n",
" NaN | \n",
" 10882.0 | \n",
" 9957.0 | \n",
" 10621.0 | \n",
" 10603.0 | \n",
" 9830 | \n",
"
\n",
" \n",
" 48 | \n",
" NaN | \n",
" NaN | \n",
" 10958.0 | \n",
" 10033.0 | \n",
" 10538.0 | \n",
" 10439.0 | \n",
" 9822 | \n",
"
\n",
" \n",
" 49 | \n",
" NaN | \n",
" NaN | \n",
" 10816.0 | \n",
" 10287.0 | \n",
" 10781.0 | \n",
" 11223.0 | \n",
" 10365 | \n",
"
\n",
" \n",
" 50 | \n",
" NaN | \n",
" NaN | \n",
" 11188.0 | \n",
" 10550.0 | \n",
" 11217.0 | \n",
" 10533.0 | \n",
" 10269 | \n",
"
\n",
" \n",
" 51 | \n",
" NaN | \n",
" NaN | \n",
" 11926.0 | \n",
" 11116.0 | \n",
" 12517.0 | \n",
" 11493.0 | \n",
" 10689 | \n",
"
\n",
" \n",
" 52 | \n",
" NaN | \n",
" NaN | \n",
" 7533.0 | \n",
" 7131.0 | \n",
" 8487.0 | \n",
" 8003.0 | \n",
" 8630 | \n",
"
\n",
" \n",
" 53 | \n",
" NaN | \n",
" NaN | \n",
" NaN | \n",
" NaN | \n",
" NaN | \n",
" NaN | \n",
" 7524 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" total_2020 total_previous total_2019 total_2018 total_2017 \\\n",
"1 12254.0 12175.0 10955.0 12723.0 11991.0 \n",
"2 14058.0 13822.0 12609.0 15050.0 13715.0 \n",
"3 12990.0 13216.0 11860.0 14256.0 13610.0 \n",
"4 11856.0 12760.0 11740.0 13935.0 12877.0 \n",
"5 11612.0 12206.0 11297.0 13285.0 12485.0 \n",
"6 10986.0 11925.0 11660.0 12495.0 12269.0 \n",
"7 10944.0 11627.0 11824.0 12246.0 11644.0 \n",
"8 10841.0 11548.0 11295.0 12142.0 11794.0 \n",
"9 10816.0 11183.0 11044.0 10854.0 11248.0 \n",
"10 10895.0 11498.0 10898.0 12997.0 11077.0 \n",
"11 11019.0 11205.0 10567.0 12788.0 10697.0 \n",
"12 10645.0 10573.0 10402.0 11913.0 10325.0 \n",
"13 11141.0 10130.0 9867.0 9941.0 10027.0 \n",
"14 16387.0 10305.0 10126.0 10794.0 9939.0 \n",
"15 18516.0 10520.0 10291.0 12301.0 8493.0 \n",
"16 22351.0 10497.0 9025.0 11223.0 9644.0 \n",
"17 21997.0 10458.0 10059.0 10306.0 10908.0 \n",
"18 NaN NaN 11207.0 10153.0 9064.0 \n",
"19 NaN NaN 9055.0 8624.0 10693.0 \n",
"20 NaN NaN 10272.0 10141.0 10288.0 \n",
"21 NaN NaN 10284.0 9636.0 10040.0 \n",
"22 NaN NaN 8260.0 8147.0 8332.0 \n",
"23 NaN NaN 10140.0 9950.0 9766.0 \n",
"24 NaN NaN 9445.0 9343.0 9367.0 \n",
"25 NaN NaN 9458.0 9256.0 9627.0 \n",
"26 NaN NaN 9511.0 9212.0 9334.0 \n",
"27 NaN NaN 9062.0 9258.0 9263.0 \n",
"28 NaN NaN 9179.0 9293.0 9376.0 \n",
"29 NaN NaN 9080.0 9127.0 9113.0 \n",
"30 NaN NaN 9112.0 9141.0 8882.0 \n",
"31 NaN NaN 9271.0 9161.0 8941.0 \n",
"32 NaN NaN 9122.0 9319.0 9038.0 \n",
"33 NaN NaN 9093.0 8830.0 9299.0 \n",
"34 NaN NaN 8994.0 8978.0 9382.0 \n",
"35 NaN NaN 8242.0 7865.0 8149.0 \n",
"36 NaN NaN 9695.0 9445.0 9497.0 \n",
"37 NaN NaN 9513.0 9191.0 9454.0 \n",
"38 NaN NaN 9440.0 9305.0 9534.0 \n",
"39 NaN NaN 9517.0 9150.0 9689.0 \n",
"40 NaN NaN 9799.0 9503.0 9778.0 \n",
"41 NaN NaN 9973.0 9649.0 9940.0 \n",
"42 NaN NaN 10156.0 9864.0 10031.0 \n",
"43 NaN NaN 10021.0 9603.0 9739.0 \n",
"44 NaN NaN 10164.0 9529.0 9984.0 \n",
"45 NaN NaN 10697.0 10151.0 10346.0 \n",
"46 NaN NaN 10650.0 10193.0 10275.0 \n",
"47 NaN NaN 10882.0 9957.0 10621.0 \n",
"48 NaN NaN 10958.0 10033.0 10538.0 \n",
"49 NaN NaN 10816.0 10287.0 10781.0 \n",
"50 NaN NaN 11188.0 10550.0 11217.0 \n",
"51 NaN NaN 11926.0 11116.0 12517.0 \n",
"52 NaN NaN 7533.0 7131.0 8487.0 \n",
"53 NaN NaN NaN NaN NaN \n",
"\n",
" total_2016 total_2015 \n",
"1 13045.0 12286 \n",
"2 11501.0 16237 \n",
"3 11473.0 14866 \n",
"4 11317.0 13934 \n",
"5 11052.0 12900 \n",
"6 11170.0 12039 \n",
"7 10590.0 11822 \n",
"8 11056.0 11434 \n",
"9 11285.0 11472 \n",
"10 11010.0 11469 \n",
"11 11022.0 10951 \n",
"12 9635.0 10568 \n",
"13 10286.0 10493 \n",
"14 11599.0 9062 \n",
"15 11417.0 10089 \n",
"16 10925.0 11639 \n",
"17 10413.0 10599 \n",
"18 9137.0 10134 \n",
"19 10637.0 8862 \n",
"20 9953.0 10290 \n",
"21 9739.0 10005 \n",
"22 7909.0 8213 \n",
"23 9873.0 10157 \n",
"24 9386.0 9548 \n",
"25 9365.0 9312 \n",
"26 9228.0 9190 \n",
"27 9138.0 9205 \n",
"28 9388.0 9015 \n",
"29 9350.0 8802 \n",
"30 9335.0 8791 \n",
"31 9182.0 8617 \n",
"32 9172.0 8862 \n",
"33 9070.0 9148 \n",
"34 9319.0 9121 \n",
"35 7923.0 9026 \n",
"36 9399.0 7878 \n",
"37 9124.0 9258 \n",
"38 8945.0 9097 \n",
"39 8994.0 9529 \n",
"40 9291.0 9410 \n",
"41 9719.0 9776 \n",
"42 9768.0 9511 \n",
"43 9724.0 9711 \n",
"44 10152.0 9618 \n",
"45 10470.0 9994 \n",
"46 10694.0 9938 \n",
"47 10603.0 9830 \n",
"48 10439.0 9822 \n",
"49 11223.0 10365 \n",
"50 10533.0 10269 \n",
"51 11493.0 10689 \n",
"52 8003.0 8630 \n",
"53 NaN 7524 "
]
},
"execution_count": 121,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"deaths_headlines = deaths_headlines.merge(dh19['total_2019'], how='outer', left_index=True, right_index=True)\n",
"deaths_headlines = deaths_headlines.merge(dh18['total_2018'], how='outer', left_index=True, right_index=True)\n",
"deaths_headlines = deaths_headlines.merge(dh17['total_2017'], how='outer', left_index=True, right_index=True)\n",
"deaths_headlines = deaths_headlines.merge(dh16['total_2016'], how='outer', left_index=True, right_index=True)\n",
"deaths_headlines = deaths_headlines.merge(dh15['total_2015'], how='outer', left_index=True, right_index=True)\n",
"deaths_headlines"
]
},
{
"cell_type": "code",
"execution_count": 162,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 162,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"