projects / covid19.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
history | raw | HEAD
General updates
[covid19.git] / excess_death_accuracy.ipynb
1 {
2 "cells": [
3 {
4 "cell_type": "code",
5 "execution_count": 17,
6 "metadata": {},
7 "outputs": [],
8 "source": [
9 "import itertools\n",
10 "import collections\n",
11 "import json\n",
12 "import pandas as pd\n",
13 "import numpy as np\n",
14 "from scipy.stats import gmean\n",
15 "import datetime\n",
16 "\n",
17 "import matplotlib as mpl\n",
18 "import matplotlib.pyplot as plt\n",
19 "%matplotlib inline"
20 ]
21 },
22 {
23 "cell_type": "code",
24 "execution_count": 18,
25 "metadata": {},
26 "outputs": [
27 {
28 "data": {
29 "text/html": [
30 "<div>\n",
31 "<style scoped>\n",
32 " .dataframe tbody tr th:only-of-type {\n",
33 " vertical-align: middle;\n",
34 " }\n",
35 "\n",
36 " .dataframe tbody tr th {\n",
37 " vertical-align: top;\n",
38 " }\n",
39 "\n",
40 " .dataframe thead th {\n",
41 " text-align: right;\n",
42 " }\n",
43 "</style>\n",
44 "<table border=\"1\" class=\"dataframe\">\n",
45 " <thead>\n",
46 " <tr style=\"text-align: right;\">\n",
47 " <th></th>\n",
48 " <th>cases</th>\n",
49 " <th>deaths</th>\n",
50 " <th>cases_culm</th>\n",
51 " <th>deaths_culm</th>\n",
52 " <th>cases_diff</th>\n",
53 " <th>deaths_diff</th>\n",
54 " </tr>\n",
55 " <tr>\n",
56 " <th>dateRep</th>\n",
57 " <th></th>\n",
58 " <th></th>\n",
59 " <th></th>\n",
60 " <th></th>\n",
61 " <th></th>\n",
62 " <th></th>\n",
63 " </tr>\n",
64 " </thead>\n",
65 " <tbody>\n",
66 " <tr>\n",
67 " <td>2019-12-31</td>\n",
68 " <td>0</td>\n",
69 " <td>0</td>\n",
70 " <td>0</td>\n",
71 " <td>0</td>\n",
72 " <td>NaN</td>\n",
73 " <td>NaN</td>\n",
74 " </tr>\n",
75 " <tr>\n",
76 " <td>2020-01-01</td>\n",
77 " <td>0</td>\n",
78 " <td>0</td>\n",
79 " <td>0</td>\n",
80 " <td>0</td>\n",
81 " <td>0.0</td>\n",
82 " <td>0.0</td>\n",
83 " </tr>\n",
84 " <tr>\n",
85 " <td>2020-01-02</td>\n",
86 " <td>0</td>\n",
87 " <td>0</td>\n",
88 " <td>0</td>\n",
89 " <td>0</td>\n",
90 " <td>0.0</td>\n",
91 " <td>0.0</td>\n",
92 " </tr>\n",
93 " <tr>\n",
94 " <td>2020-01-03</td>\n",
95 " <td>0</td>\n",
96 " <td>0</td>\n",
97 " <td>0</td>\n",
98 " <td>0</td>\n",
99 " <td>0.0</td>\n",
100 " <td>0.0</td>\n",
101 " </tr>\n",
102 " <tr>\n",
103 " <td>2020-01-04</td>\n",
104 " <td>0</td>\n",
105 " <td>0</td>\n",
106 " <td>0</td>\n",
107 " <td>0</td>\n",
108 " <td>0.0</td>\n",
109 " <td>0.0</td>\n",
110 " </tr>\n",
111 " <tr>\n",
112 " <td>...</td>\n",
113 " <td>...</td>\n",
114 " <td>...</td>\n",
115 " <td>...</td>\n",
116 " <td>...</td>\n",
117 " <td>...</td>\n",
118 " <td>...</td>\n",
119 " </tr>\n",
120 " <tr>\n",
121 " <td>2020-08-08</td>\n",
122 " <td>871</td>\n",
123 " <td>98</td>\n",
124 " <td>309005</td>\n",
125 " <td>46511</td>\n",
126 " <td>-79.0</td>\n",
127 " <td>49.0</td>\n",
128 " </tr>\n",
129 " <tr>\n",
130 " <td>2020-08-09</td>\n",
131 " <td>758</td>\n",
132 " <td>55</td>\n",
133 " <td>309763</td>\n",
134 " <td>46566</td>\n",
135 " <td>-113.0</td>\n",
136 " <td>-43.0</td>\n",
137 " </tr>\n",
138 " <tr>\n",
139 " <td>2020-08-10</td>\n",
140 " <td>1062</td>\n",
141 " <td>8</td>\n",
142 " <td>310825</td>\n",
143 " <td>46574</td>\n",
144 " <td>304.0</td>\n",
145 " <td>-47.0</td>\n",
146 " </tr>\n",
147 " <tr>\n",
148 " <td>2020-08-11</td>\n",
149 " <td>816</td>\n",
150 " <td>-48</td>\n",
151 " <td>311641</td>\n",
152 " <td>46526</td>\n",
153 " <td>-246.0</td>\n",
154 " <td>-56.0</td>\n",
155 " </tr>\n",
156 " <tr>\n",
157 " <td>2020-08-12</td>\n",
158 " <td>1148</td>\n",
159 " <td>0</td>\n",
160 " <td>312789</td>\n",
161 " <td>46526</td>\n",
162 " <td>332.0</td>\n",
163 " <td>48.0</td>\n",
164 " </tr>\n",
165 " </tbody>\n",
166 "</table>\n",
167 "<p>226 rows × 6 columns</p>\n",
168 "</div>"
169 ],
170 "text/plain": [
171 " cases deaths cases_culm deaths_culm cases_diff deaths_diff\n",
172 "dateRep \n",
173 "2019-12-31 0 0 0 0 NaN NaN\n",
174 "2020-01-01 0 0 0 0 0.0 0.0\n",
175 "2020-01-02 0 0 0 0 0.0 0.0\n",
176 "2020-01-03 0 0 0 0 0.0 0.0\n",
177 "2020-01-04 0 0 0 0 0.0 0.0\n",
178 "... ... ... ... ... ... ...\n",
179 "2020-08-08 871 98 309005 46511 -79.0 49.0\n",
180 "2020-08-09 758 55 309763 46566 -113.0 -43.0\n",
181 "2020-08-10 1062 8 310825 46574 304.0 -47.0\n",
182 "2020-08-11 816 -48 311641 46526 -246.0 -56.0\n",
183 "2020-08-12 1148 0 312789 46526 332.0 48.0\n",
184 "\n",
185 "[226 rows x 6 columns]"
186 ]
187 },
188 "execution_count": 18,
189 "metadata": {},
190 "output_type": "execute_result"
191 }
192 ],
193 "source": [
194 "data_by_day = pd.read_csv('data_by_day_uk.csv', index_col='dateRep', parse_dates=True)\n",
195 "data_by_day"
196 ]
197 },
198 {
199 "cell_type": "code",
200 "execution_count": 19,
201 "metadata": {},
202 "outputs": [
203 {
204 "data": {
205 "text/html": [
206 "<div>\n",
207 "<style scoped>\n",
208 " .dataframe tbody tr th:only-of-type {\n",
209 " vertical-align: middle;\n",
210 " }\n",
211 "\n",
212 " .dataframe tbody tr th {\n",
213 " vertical-align: top;\n",
214 " }\n",
215 "\n",
216 " .dataframe thead th {\n",
217 " text-align: right;\n",
218 " }\n",
219 "</style>\n",
220 "<table border=\"1\" class=\"dataframe\">\n",
221 " <thead>\n",
222 " <tr style=\"text-align: right;\">\n",
223 " <th></th>\n",
224 " <th>total_2020</th>\n",
225 " <th>total_2019</th>\n",
226 " <th>total_2018</th>\n",
227 " <th>total_2017</th>\n",
228 " <th>total_2016</th>\n",
229 " <th>total_2015</th>\n",
230 " <th>previous_mean</th>\n",
231 " </tr>\n",
232 " <tr>\n",
233 " <th>week_ended</th>\n",
234 " <th></th>\n",
235 " <th></th>\n",
236 " <th></th>\n",
237 " <th></th>\n",
238 " <th></th>\n",
239 " <th></th>\n",
240 " <th></th>\n",
241 " </tr>\n",
242 " </thead>\n",
243 " <tbody>\n",
244 " <tr>\n",
245 " <td>2020-01-03</td>\n",
246 " <td>13768.0</td>\n",
247 " <td>12424.0</td>\n",
248 " <td>14701.0</td>\n",
249 " <td>13612.0</td>\n",
250 " <td>14863.0</td>\n",
251 " <td>13751</td>\n",
252 " <td>13870.2</td>\n",
253 " </tr>\n",
254 " <tr>\n",
255 " <td>2020-01-10</td>\n",
256 " <td>16020.0</td>\n",
257 " <td>14487.0</td>\n",
258 " <td>17430.0</td>\n",
259 " <td>15528.0</td>\n",
260 " <td>13154.0</td>\n",
261 " <td>18318</td>\n",
262 " <td>15783.4</td>\n",
263 " </tr>\n",
264 " <tr>\n",
265 " <td>2020-01-17</td>\n",
266 " <td>14723.0</td>\n",
267 " <td>13545.0</td>\n",
268 " <td>16355.0</td>\n",
269 " <td>15231.0</td>\n",
270 " <td>13060.0</td>\n",
271 " <td>16738</td>\n",
272 " <td>14985.8</td>\n",
273 " </tr>\n",
274 " <tr>\n",
275 " <td>2020-01-24</td>\n",
276 " <td>13429.0</td>\n",
277 " <td>13283.0</td>\n",
278 " <td>15971.0</td>\n",
279 " <td>14461.0</td>\n",
280 " <td>12859.0</td>\n",
281 " <td>15712</td>\n",
282 " <td>14457.2</td>\n",
283 " </tr>\n",
284 " <tr>\n",
285 " <td>2020-01-31</td>\n",
286 " <td>13123.0</td>\n",
287 " <td>12799.0</td>\n",
288 " <td>15087.0</td>\n",
289 " <td>14188.0</td>\n",
290 " <td>12571.0</td>\n",
291 " <td>14560</td>\n",
292 " <td>13841.0</td>\n",
293 " </tr>\n",
294 " <tr>\n",
295 " <td>2020-02-07</td>\n",
296 " <td>12534.0</td>\n",
297 " <td>13222.0</td>\n",
298 " <td>14111.0</td>\n",
299 " <td>13805.0</td>\n",
300 " <td>12697.0</td>\n",
301 " <td>13730</td>\n",
302 " <td>13513.0</td>\n",
303 " </tr>\n",
304 " <tr>\n",
305 " <td>2020-02-14</td>\n",
306 " <td>12412.0</td>\n",
307 " <td>13347.0</td>\n",
308 " <td>13925.0</td>\n",
309 " <td>13212.0</td>\n",
310 " <td>12016.0</td>\n",
311 " <td>13510</td>\n",
312 " <td>13202.0</td>\n",
313 " </tr>\n",
314 " <tr>\n",
315 " <td>2020-02-21</td>\n",
316 " <td>12300.0</td>\n",
317 " <td>12877.0</td>\n",
318 " <td>13753.0</td>\n",
319 " <td>13330.0</td>\n",
320 " <td>12718.0</td>\n",
321 " <td>13071</td>\n",
322 " <td>13149.8</td>\n",
323 " </tr>\n",
324 " <tr>\n",
325 " <td>2020-02-28</td>\n",
326 " <td>12334.0</td>\n",
327 " <td>12479.0</td>\n",
328 " <td>12190.0</td>\n",
329 " <td>12819.0</td>\n",
330 " <td>12733.0</td>\n",
331 " <td>13181</td>\n",
332 " <td>12680.4</td>\n",
333 " </tr>\n",
334 " <tr>\n",
335 " <td>2020-03-06</td>\n",
336 " <td>12415.0</td>\n",
337 " <td>12396.0</td>\n",
338 " <td>14859.0</td>\n",
339 " <td>12580.0</td>\n",
340 " <td>12493.0</td>\n",
341 " <td>13007</td>\n",
342 " <td>13067.0</td>\n",
343 " </tr>\n",
344 " <tr>\n",
345 " <td>2020-03-13</td>\n",
346 " <td>12499.0</td>\n",
347 " <td>12018.0</td>\n",
348 " <td>14367.0</td>\n",
349 " <td>12089.0</td>\n",
350 " <td>12489.0</td>\n",
351 " <td>12475</td>\n",
352 " <td>12687.6</td>\n",
353 " </tr>\n",
354 " <tr>\n",
355 " <td>2020-03-20</td>\n",
356 " <td>12112.0</td>\n",
357 " <td>11797.0</td>\n",
358 " <td>13397.0</td>\n",
359 " <td>11833.0</td>\n",
360 " <td>10983.0</td>\n",
361 " <td>12027</td>\n",
362 " <td>12007.4</td>\n",
363 " </tr>\n",
364 " <tr>\n",
365 " <td>2020-03-27</td>\n",
366 " <td>12507.0</td>\n",
367 " <td>11260.0</td>\n",
368 " <td>11310.0</td>\n",
369 " <td>11453.0</td>\n",
370 " <td>11738.0</td>\n",
371 " <td>11987</td>\n",
372 " <td>11549.6</td>\n",
373 " </tr>\n",
374 " <tr>\n",
375 " <td>2020-04-03</td>\n",
376 " <td>18565.0</td>\n",
377 " <td>11445.0</td>\n",
378 " <td>12272.0</td>\n",
379 " <td>11305.0</td>\n",
380 " <td>13060.0</td>\n",
381 " <td>10325</td>\n",
382 " <td>11681.4</td>\n",
383 " </tr>\n",
384 " <tr>\n",
385 " <td>2020-04-10</td>\n",
386 " <td>20929.0</td>\n",
387 " <td>11661.0</td>\n",
388 " <td>13843.0</td>\n",
389 " <td>9761.0</td>\n",
390 " <td>12757.0</td>\n",
391 " <td>11575</td>\n",
392 " <td>11919.4</td>\n",
393 " </tr>\n",
394 " <tr>\n",
395 " <td>2020-04-17</td>\n",
396 " <td>24691.0</td>\n",
397 " <td>10243.0</td>\n",
398 " <td>12639.0</td>\n",
399 " <td>11000.0</td>\n",
400 " <td>12310.0</td>\n",
401 " <td>13061</td>\n",
402 " <td>11850.6</td>\n",
403 " </tr>\n",
404 " <tr>\n",
405 " <td>2020-04-24</td>\n",
406 " <td>24303.0</td>\n",
407 " <td>11452.0</td>\n",
408 " <td>11596.0</td>\n",
409 " <td>12356.0</td>\n",
410 " <td>11795.0</td>\n",
411 " <td>12023</td>\n",
412 " <td>11844.4</td>\n",
413 " </tr>\n",
414 " <tr>\n",
415 " <td>2020-05-01</td>\n",
416 " <td>20059.0</td>\n",
417 " <td>12695.0</td>\n",
418 " <td>11538.0</td>\n",
419 " <td>10372.0</td>\n",
420 " <td>10401.0</td>\n",
421 " <td>11586</td>\n",
422 " <td>11318.4</td>\n",
423 " </tr>\n",
424 " <tr>\n",
425 " <td>2020-05-08</td>\n",
426 " <td>14428.0</td>\n",
427 " <td>10361.0</td>\n",
428 " <td>9821.0</td>\n",
429 " <td>12114.0</td>\n",
430 " <td>12002.0</td>\n",
431 " <td>10138</td>\n",
432 " <td>10887.2</td>\n",
433 " </tr>\n",
434 " <tr>\n",
435 " <td>2020-05-15</td>\n",
436 " <td>16390.0</td>\n",
437 " <td>11717.0</td>\n",
438 " <td>11386.0</td>\n",
439 " <td>11718.0</td>\n",
440 " <td>11222.0</td>\n",
441 " <td>11692</td>\n",
442 " <td>11547.0</td>\n",
443 " </tr>\n",
444 " <tr>\n",
445 " <td>2020-05-22</td>\n",
446 " <td>13839.0</td>\n",
447 " <td>11653.0</td>\n",
448 " <td>10974.0</td>\n",
449 " <td>11431.0</td>\n",
450 " <td>11013.0</td>\n",
451 " <td>11334</td>\n",
452 " <td>11281.0</td>\n",
453 " </tr>\n",
454 " <tr>\n",
455 " <td>2020-05-29</td>\n",
456 " <td>11265.0</td>\n",
457 " <td>9534.0</td>\n",
458 " <td>9397.0</td>\n",
459 " <td>9603.0</td>\n",
460 " <td>9192.0</td>\n",
461 " <td>9514</td>\n",
462 " <td>9448.0</td>\n",
463 " </tr>\n",
464 " <tr>\n",
465 " <td>2020-06-05</td>\n",
466 " <td>12106.0</td>\n",
467 " <td>11461.0</td>\n",
468 " <td>11259.0</td>\n",
469 " <td>11134.0</td>\n",
470 " <td>11171.0</td>\n",
471 " <td>11603</td>\n",
472 " <td>11325.6</td>\n",
473 " </tr>\n",
474 " <tr>\n",
475 " <td>2020-06-12</td>\n",
476 " <td>11302.0</td>\n",
477 " <td>10754.0</td>\n",
478 " <td>10535.0</td>\n",
479 " <td>10698.0</td>\n",
480 " <td>10673.0</td>\n",
481 " <td>10858</td>\n",
482 " <td>10703.6</td>\n",
483 " </tr>\n",
484 " <tr>\n",
485 " <td>2020-06-19</td>\n",
486 " <td>10694.0</td>\n",
487 " <td>10807.0</td>\n",
488 " <td>10514.0</td>\n",
489 " <td>10930.0</td>\n",
490 " <td>10611.0</td>\n",
491 " <td>10629</td>\n",
492 " <td>10698.2</td>\n",
493 " </tr>\n",
494 " <tr>\n",
495 " <td>2020-06-26</td>\n",
496 " <td>10282.0</td>\n",
497 " <td>10824.0</td>\n",
498 " <td>10529.0</td>\n",
499 " <td>10624.0</td>\n",
500 " <td>10526.0</td>\n",
501 " <td>10525</td>\n",
502 " <td>10605.6</td>\n",
503 " </tr>\n",
504 " <tr>\n",
505 " <td>2020-07-03</td>\n",
506 " <td>10412.0</td>\n",
507 " <td>10328.0</td>\n",
508 " <td>10565.0</td>\n",
509 " <td>10565.0</td>\n",
510 " <td>10412.0</td>\n",
511 " <td>10545</td>\n",
512 " <td>10483.0</td>\n",
513 " </tr>\n",
514 " <tr>\n",
515 " <td>2020-07-10</td>\n",
516 " <td>9941.0</td>\n",
517 " <td>10512.0</td>\n",
518 " <td>10467.0</td>\n",
519 " <td>10643.0</td>\n",
520 " <td>10647.0</td>\n",
521 " <td>10278</td>\n",
522 " <td>10509.4</td>\n",
523 " </tr>\n",
524 " <tr>\n",
525 " <td>2020-07-17</td>\n",
526 " <td>10096.0</td>\n",
527 " <td>10324.0</td>\n",
528 " <td>10353.0</td>\n",
529 " <td>10426.0</td>\n",
530 " <td>10672.0</td>\n",
531 " <td>10028</td>\n",
532 " <td>10360.6</td>\n",
533 " </tr>\n",
534 " <tr>\n",
535 " <td>2020-07-24</td>\n",
536 " <td>10159.0</td>\n",
537 " <td>10422.0</td>\n",
538 " <td>10356.0</td>\n",
539 " <td>10147.0</td>\n",
540 " <td>10612.0</td>\n",
541 " <td>10021</td>\n",
542 " <td>10311.6</td>\n",
543 " </tr>\n",
544 " <tr>\n",
545 " <td>2020-07-31</td>\n",
546 " <td>10262.0</td>\n",
547 " <td>10564.0</td>\n",
548 " <td>10408.0</td>\n",
549 " <td>10239.0</td>\n",
550 " <td>10433.0</td>\n",
551 " <td>9893</td>\n",
552 " <td>10307.4</td>\n",
553 " </tr>\n",
554 " <tr>\n",
555 " <td>2020-08-07</td>\n",
556 " <td>NaN</td>\n",
557 " <td>10406.0</td>\n",
558 " <td>10542.0</td>\n",
559 " <td>10278.0</td>\n",
560 " <td>10439.0</td>\n",
561 " <td>10153</td>\n",
562 " <td>10363.6</td>\n",
563 " </tr>\n",
564 " <tr>\n",
565 " <td>2020-08-14</td>\n",
566 " <td>NaN</td>\n",
567 " <td>10405.0</td>\n",
568 " <td>10091.0</td>\n",
569 " <td>10569.0</td>\n",
570 " <td>10312.0</td>\n",
571 " <td>10352</td>\n",
572 " <td>10345.8</td>\n",
573 " </tr>\n",
574 " <tr>\n",
575 " <td>2020-08-21</td>\n",
576 " <td>NaN</td>\n",
577 " <td>10279.0</td>\n",
578 " <td>10199.0</td>\n",
579 " <td>10698.0</td>\n",
580 " <td>10637.0</td>\n",
581 " <td>10354</td>\n",
582 " <td>10433.4</td>\n",
583 " </tr>\n",
584 " <tr>\n",
585 " <td>2020-08-28</td>\n",
586 " <td>NaN</td>\n",
587 " <td>9478.0</td>\n",
588 " <td>9046.0</td>\n",
589 " <td>9372.0</td>\n",
590 " <td>9226.0</td>\n",
591 " <td>10239</td>\n",
592 " <td>9472.2</td>\n",
593 " </tr>\n",
594 " <tr>\n",
595 " <td>2020-09-04</td>\n",
596 " <td>NaN</td>\n",
597 " <td>10918.0</td>\n",
598 " <td>10680.0</td>\n",
599 " <td>10781.0</td>\n",
600 " <td>10681.0</td>\n",
601 " <td>9092</td>\n",
602 " <td>10430.4</td>\n",
603 " </tr>\n",
604 " <tr>\n",
605 " <td>2020-09-11</td>\n",
606 " <td>NaN</td>\n",
607 " <td>10892.0</td>\n",
608 " <td>10496.0</td>\n",
609 " <td>10692.0</td>\n",
610 " <td>10401.0</td>\n",
611 " <td>10573</td>\n",
612 " <td>10610.8</td>\n",
613 " </tr>\n",
614 " <tr>\n",
615 " <td>2020-09-18</td>\n",
616 " <td>NaN</td>\n",
617 " <td>10792.0</td>\n",
618 " <td>10498.0</td>\n",
619 " <td>10875.0</td>\n",
620 " <td>10183.0</td>\n",
621 " <td>10381</td>\n",
622 " <td>10545.8</td>\n",
623 " </tr>\n",
624 " <tr>\n",
625 " <td>2020-09-25</td>\n",
626 " <td>NaN</td>\n",
627 " <td>10954.0</td>\n",
628 " <td>10463.0</td>\n",
629 " <td>11027.0</td>\n",
630 " <td>10278.0</td>\n",
631 " <td>10826</td>\n",
632 " <td>10709.6</td>\n",
633 " </tr>\n",
634 " <tr>\n",
635 " <td>2020-10-02</td>\n",
636 " <td>NaN</td>\n",
637 " <td>11113.0</td>\n",
638 " <td>10869.0</td>\n",
639 " <td>11101.0</td>\n",
640 " <td>10671.0</td>\n",
641 " <td>10700</td>\n",
642 " <td>10890.8</td>\n",
643 " </tr>\n",
644 " <tr>\n",
645 " <td>2020-10-09</td>\n",
646 " <td>NaN</td>\n",
647 " <td>11403.0</td>\n",
648 " <td>11048.0</td>\n",
649 " <td>11357.0</td>\n",
650 " <td>11016.0</td>\n",
651 " <td>11108</td>\n",
652 " <td>11186.4</td>\n",
653 " </tr>\n",
654 " <tr>\n",
655 " <td>2020-10-16</td>\n",
656 " <td>NaN</td>\n",
657 " <td>11625.0</td>\n",
658 " <td>11177.0</td>\n",
659 " <td>11389.0</td>\n",
660 " <td>11134.0</td>\n",
661 " <td>10799</td>\n",
662 " <td>11224.8</td>\n",
663 " </tr>\n",
664 " <tr>\n",
665 " <td>2020-10-23</td>\n",
666 " <td>NaN</td>\n",
667 " <td>11415.0</td>\n",
668 " <td>10885.0</td>\n",
669 " <td>11152.0</td>\n",
670 " <td>11048.0</td>\n",
671 " <td>10966</td>\n",
672 " <td>11093.2</td>\n",
673 " </tr>\n",
674 " <tr>\n",
675 " <td>2020-10-30</td>\n",
676 " <td>NaN</td>\n",
677 " <td>11567.0</td>\n",
678 " <td>10866.0</td>\n",
679 " <td>11366.0</td>\n",
680 " <td>11463.0</td>\n",
681 " <td>11026</td>\n",
682 " <td>11257.6</td>\n",
683 " </tr>\n",
684 " <tr>\n",
685 " <td>2020-11-06</td>\n",
686 " <td>NaN</td>\n",
687 " <td>12177.0</td>\n",
688 " <td>11588.0</td>\n",
689 " <td>11767.0</td>\n",
690 " <td>11803.0</td>\n",
691 " <td>11312</td>\n",
692 " <td>11729.4</td>\n",
693 " </tr>\n",
694 " <tr>\n",
695 " <td>2020-11-13</td>\n",
696 " <td>NaN</td>\n",
697 " <td>12146.0</td>\n",
698 " <td>11552.0</td>\n",
699 " <td>11773.0</td>\n",
700 " <td>12209.0</td>\n",
701 " <td>11338</td>\n",
702 " <td>11803.6</td>\n",
703 " </tr>\n",
704 " <tr>\n",
705 " <td>2020-11-20</td>\n",
706 " <td>NaN</td>\n",
707 " <td>12472.0</td>\n",
708 " <td>11289.0</td>\n",
709 " <td>12102.0</td>\n",
710 " <td>12064.0</td>\n",
711 " <td>11178</td>\n",
712 " <td>11821.0</td>\n",
713 " </tr>\n",
714 " <tr>\n",
715 " <td>2020-11-27</td>\n",
716 " <td>NaN</td>\n",
717 " <td>12455.0</td>\n",
718 " <td>11392.0</td>\n",
719 " <td>12046.0</td>\n",
720 " <td>11901.0</td>\n",
721 " <td>11216</td>\n",
722 " <td>11802.0</td>\n",
723 " </tr>\n",
724 " <tr>\n",
725 " <td>2020-12-04</td>\n",
726 " <td>NaN</td>\n",
727 " <td>12275.0</td>\n",
728 " <td>11687.0</td>\n",
729 " <td>12342.0</td>\n",
730 " <td>12733.0</td>\n",
731 " <td>11748</td>\n",
732 " <td>12157.0</td>\n",
733 " </tr>\n",
734 " <tr>\n",
735 " <td>2020-12-11</td>\n",
736 " <td>NaN</td>\n",
737 " <td>12853.0</td>\n",
738 " <td>12078.0</td>\n",
739 " <td>12924.0</td>\n",
740 " <td>12076.0</td>\n",
741 " <td>11713</td>\n",
742 " <td>12328.8</td>\n",
743 " </tr>\n",
744 " <tr>\n",
745 " <td>2020-12-18</td>\n",
746 " <td>NaN</td>\n",
747 " <td>13566.0</td>\n",
748 " <td>12649.0</td>\n",
749 " <td>14308.0</td>\n",
750 " <td>13137.0</td>\n",
751 " <td>12136</td>\n",
752 " <td>13159.2</td>\n",
753 " </tr>\n",
754 " <tr>\n",
755 " <td>2020-12-25</td>\n",
756 " <td>NaN</td>\n",
757 " <td>8727.0</td>\n",
758 " <td>8384.0</td>\n",
759 " <td>9904.0</td>\n",
760 " <td>9335.0</td>\n",
761 " <td>9806</td>\n",
762 " <td>9231.2</td>\n",
763 " </tr>\n",
764 " </tbody>\n",
765 "</table>\n",
766 "</div>"
767 ],
768 "text/plain": [
769 " total_2020 total_2019 total_2018 total_2017 total_2016 \\\n",
770 "week_ended \n",
771 "2020-01-03 13768.0 12424.0 14701.0 13612.0 14863.0 \n",
772 "2020-01-10 16020.0 14487.0 17430.0 15528.0 13154.0 \n",
773 "2020-01-17 14723.0 13545.0 16355.0 15231.0 13060.0 \n",
774 "2020-01-24 13429.0 13283.0 15971.0 14461.0 12859.0 \n",
775 "2020-01-31 13123.0 12799.0 15087.0 14188.0 12571.0 \n",
776 "2020-02-07 12534.0 13222.0 14111.0 13805.0 12697.0 \n",
777 "2020-02-14 12412.0 13347.0 13925.0 13212.0 12016.0 \n",
778 "2020-02-21 12300.0 12877.0 13753.0 13330.0 12718.0 \n",
779 "2020-02-28 12334.0 12479.0 12190.0 12819.0 12733.0 \n",
780 "2020-03-06 12415.0 12396.0 14859.0 12580.0 12493.0 \n",
781 "2020-03-13 12499.0 12018.0 14367.0 12089.0 12489.0 \n",
782 "2020-03-20 12112.0 11797.0 13397.0 11833.0 10983.0 \n",
783 "2020-03-27 12507.0 11260.0 11310.0 11453.0 11738.0 \n",
784 "2020-04-03 18565.0 11445.0 12272.0 11305.0 13060.0 \n",
785 "2020-04-10 20929.0 11661.0 13843.0 9761.0 12757.0 \n",
786 "2020-04-17 24691.0 10243.0 12639.0 11000.0 12310.0 \n",
787 "2020-04-24 24303.0 11452.0 11596.0 12356.0 11795.0 \n",
788 "2020-05-01 20059.0 12695.0 11538.0 10372.0 10401.0 \n",
789 "2020-05-08 14428.0 10361.0 9821.0 12114.0 12002.0 \n",
790 "2020-05-15 16390.0 11717.0 11386.0 11718.0 11222.0 \n",
791 "2020-05-22 13839.0 11653.0 10974.0 11431.0 11013.0 \n",
792 "2020-05-29 11265.0 9534.0 9397.0 9603.0 9192.0 \n",
793 "2020-06-05 12106.0 11461.0 11259.0 11134.0 11171.0 \n",
794 "2020-06-12 11302.0 10754.0 10535.0 10698.0 10673.0 \n",
795 "2020-06-19 10694.0 10807.0 10514.0 10930.0 10611.0 \n",
796 "2020-06-26 10282.0 10824.0 10529.0 10624.0 10526.0 \n",
797 "2020-07-03 10412.0 10328.0 10565.0 10565.0 10412.0 \n",
798 "2020-07-10 9941.0 10512.0 10467.0 10643.0 10647.0 \n",
799 "2020-07-17 10096.0 10324.0 10353.0 10426.0 10672.0 \n",
800 "2020-07-24 10159.0 10422.0 10356.0 10147.0 10612.0 \n",
801 "2020-07-31 10262.0 10564.0 10408.0 10239.0 10433.0 \n",
802 "2020-08-07 NaN 10406.0 10542.0 10278.0 10439.0 \n",
803 "2020-08-14 NaN 10405.0 10091.0 10569.0 10312.0 \n",
804 "2020-08-21 NaN 10279.0 10199.0 10698.0 10637.0 \n",
805 "2020-08-28 NaN 9478.0 9046.0 9372.0 9226.0 \n",
806 "2020-09-04 NaN 10918.0 10680.0 10781.0 10681.0 \n",
807 "2020-09-11 NaN 10892.0 10496.0 10692.0 10401.0 \n",
808 "2020-09-18 NaN 10792.0 10498.0 10875.0 10183.0 \n",
809 "2020-09-25 NaN 10954.0 10463.0 11027.0 10278.0 \n",
810 "2020-10-02 NaN 11113.0 10869.0 11101.0 10671.0 \n",
811 "2020-10-09 NaN 11403.0 11048.0 11357.0 11016.0 \n",
812 "2020-10-16 NaN 11625.0 11177.0 11389.0 11134.0 \n",
813 "2020-10-23 NaN 11415.0 10885.0 11152.0 11048.0 \n",
814 "2020-10-30 NaN 11567.0 10866.0 11366.0 11463.0 \n",
815 "2020-11-06 NaN 12177.0 11588.0 11767.0 11803.0 \n",
816 "2020-11-13 NaN 12146.0 11552.0 11773.0 12209.0 \n",
817 "2020-11-20 NaN 12472.0 11289.0 12102.0 12064.0 \n",
818 "2020-11-27 NaN 12455.0 11392.0 12046.0 11901.0 \n",
819 "2020-12-04 NaN 12275.0 11687.0 12342.0 12733.0 \n",
820 "2020-12-11 NaN 12853.0 12078.0 12924.0 12076.0 \n",
821 "2020-12-18 NaN 13566.0 12649.0 14308.0 13137.0 \n",
822 "2020-12-25 NaN 8727.0 8384.0 9904.0 9335.0 \n",
823 "\n",
824 " total_2015 previous_mean \n",
825 "week_ended \n",
826 "2020-01-03 13751 13870.2 \n",
827 "2020-01-10 18318 15783.4 \n",
828 "2020-01-17 16738 14985.8 \n",
829 "2020-01-24 15712 14457.2 \n",
830 "2020-01-31 14560 13841.0 \n",
831 "2020-02-07 13730 13513.0 \n",
832 "2020-02-14 13510 13202.0 \n",
833 "2020-02-21 13071 13149.8 \n",
834 "2020-02-28 13181 12680.4 \n",
835 "2020-03-06 13007 13067.0 \n",
836 "2020-03-13 12475 12687.6 \n",
837 "2020-03-20 12027 12007.4 \n",
838 "2020-03-27 11987 11549.6 \n",
839 "2020-04-03 10325 11681.4 \n",
840 "2020-04-10 11575 11919.4 \n",
841 "2020-04-17 13061 11850.6 \n",
842 "2020-04-24 12023 11844.4 \n",
843 "2020-05-01 11586 11318.4 \n",
844 "2020-05-08 10138 10887.2 \n",
845 "2020-05-15 11692 11547.0 \n",
846 "2020-05-22 11334 11281.0 \n",
847 "2020-05-29 9514 9448.0 \n",
848 "2020-06-05 11603 11325.6 \n",
849 "2020-06-12 10858 10703.6 \n",
850 "2020-06-19 10629 10698.2 \n",
851 "2020-06-26 10525 10605.6 \n",
852 "2020-07-03 10545 10483.0 \n",
853 "2020-07-10 10278 10509.4 \n",
854 "2020-07-17 10028 10360.6 \n",
855 "2020-07-24 10021 10311.6 \n",
856 "2020-07-31 9893 10307.4 \n",
857 "2020-08-07 10153 10363.6 \n",
858 "2020-08-14 10352 10345.8 \n",
859 "2020-08-21 10354 10433.4 \n",
860 "2020-08-28 10239 9472.2 \n",
861 "2020-09-04 9092 10430.4 \n",
862 "2020-09-11 10573 10610.8 \n",
863 "2020-09-18 10381 10545.8 \n",
864 "2020-09-25 10826 10709.6 \n",
865 "2020-10-02 10700 10890.8 \n",
866 "2020-10-09 11108 11186.4 \n",
867 "2020-10-16 10799 11224.8 \n",
868 "2020-10-23 10966 11093.2 \n",
869 "2020-10-30 11026 11257.6 \n",
870 "2020-11-06 11312 11729.4 \n",
871 "2020-11-13 11338 11803.6 \n",
872 "2020-11-20 11178 11821.0 \n",
873 "2020-11-27 11216 11802.0 \n",
874 "2020-12-04 11748 12157.0 \n",
875 "2020-12-11 11713 12328.8 \n",
876 "2020-12-18 12136 13159.2 \n",
877 "2020-12-25 9806 9231.2 "
878 ]
879 },
880 "execution_count": 19,
881 "metadata": {},
882 "output_type": "execute_result"
883 }
884 ],
885 "source": [
886 "deaths_by_week = pd.read_csv('deaths_by_week.csv', index_col='week_ended', parse_dates=True)\n",
887 "deaths_by_week"
888 ]
889 },
890 {
891 "cell_type": "code",
892 "execution_count": 20,
893 "metadata": {},
894 "outputs": [
895 {
896 "data": {
897 "text/html": [
898 "<div>\n",
899 "<style scoped>\n",
900 " .dataframe tbody tr th:only-of-type {\n",
901 " vertical-align: middle;\n",
902 " }\n",
903 "\n",
904 " .dataframe tbody tr th {\n",
905 " vertical-align: top;\n",
906 " }\n",
907 "\n",
908 " .dataframe thead th {\n",
909 " text-align: right;\n",
910 " }\n",
911 "</style>\n",
912 "<table border=\"1\" class=\"dataframe\">\n",
913 " <thead>\n",
914 " <tr style=\"text-align: right;\">\n",
915 " <th></th>\n",
916 " <th>cases</th>\n",
917 " <th>deaths</th>\n",
918 " <th>cases_culm</th>\n",
919 " <th>deaths_culm</th>\n",
920 " <th>cases_diff</th>\n",
921 " <th>deaths_diff</th>\n",
922 " </tr>\n",
923 " <tr>\n",
924 " <th>dateRep</th>\n",
925 " <th></th>\n",
926 " <th></th>\n",
927 " <th></th>\n",
928 " <th></th>\n",
929 " <th></th>\n",
930 " <th></th>\n",
931 " </tr>\n",
932 " </thead>\n",
933 " <tbody>\n",
934 " <tr>\n",
935 " <td>2020-01-03</td>\n",
936 " <td>0</td>\n",
937 " <td>0</td>\n",
938 " <td>0</td>\n",
939 " <td>0</td>\n",
940 " <td>0.0</td>\n",
941 " <td>0.0</td>\n",
942 " </tr>\n",
943 " <tr>\n",
944 " <td>2020-01-10</td>\n",
945 " <td>0</td>\n",
946 " <td>0</td>\n",
947 " <td>0</td>\n",
948 " <td>0</td>\n",
949 " <td>0.0</td>\n",
950 " <td>0.0</td>\n",
951 " </tr>\n",
952 " <tr>\n",
953 " <td>2020-01-17</td>\n",
954 " <td>0</td>\n",
955 " <td>0</td>\n",
956 " <td>0</td>\n",
957 " <td>0</td>\n",
958 " <td>0.0</td>\n",
959 " <td>0.0</td>\n",
960 " </tr>\n",
961 " <tr>\n",
962 " <td>2020-01-24</td>\n",
963 " <td>0</td>\n",
964 " <td>0</td>\n",
965 " <td>0</td>\n",
966 " <td>0</td>\n",
967 " <td>0.0</td>\n",
968 " <td>0.0</td>\n",
969 " </tr>\n",
970 " <tr>\n",
971 " <td>2020-01-31</td>\n",
972 " <td>0</td>\n",
973 " <td>0</td>\n",
974 " <td>0</td>\n",
975 " <td>0</td>\n",
976 " <td>0.0</td>\n",
977 " <td>0.0</td>\n",
978 " </tr>\n",
979 " <tr>\n",
980 " <td>2020-02-07</td>\n",
981 " <td>4</td>\n",
982 " <td>0</td>\n",
983 " <td>18</td>\n",
984 " <td>0</td>\n",
985 " <td>1.0</td>\n",
986 " <td>0.0</td>\n",
987 " </tr>\n",
988 " <tr>\n",
989 " <td>2020-02-14</td>\n",
990 " <td>6</td>\n",
991 " <td>0</td>\n",
992 " <td>54</td>\n",
993 " <td>0</td>\n",
994 " <td>-1.0</td>\n",
995 " <td>0.0</td>\n",
996 " </tr>\n",
997 " <tr>\n",
998 " <td>2020-02-21</td>\n",
999 " <td>0</td>\n",
1000 " <td>0</td>\n",
1001 " <td>70</td>\n",
1002 " <td>0</td>\n",
1003 " <td>0.0</td>\n",
1004 " <td>0.0</td>\n",
1005 " </tr>\n",
1006 " <tr>\n",
1007 " <td>2020-02-28</td>\n",
1008 " <td>12</td>\n",
1009 " <td>0</td>\n",
1010 " <td>96</td>\n",
1011 " <td>0</td>\n",
1012 " <td>4.0</td>\n",
1013 " <td>0.0</td>\n",
1014 " </tr>\n",
1015 " <tr>\n",
1016 " <td>2020-03-06</td>\n",
1017 " <td>198</td>\n",
1018 " <td>0</td>\n",
1019 " <td>681</td>\n",
1020 " <td>0</td>\n",
1021 " <td>52.0</td>\n",
1022 " <td>0.0</td>\n",
1023 " </tr>\n",
1024 " <tr>\n",
1025 " <td>2020-03-13</td>\n",
1026 " <td>1062</td>\n",
1027 " <td>9</td>\n",
1028 " <td>4279</td>\n",
1029 " <td>31</td>\n",
1030 " <td>350.0</td>\n",
1031 " <td>2.0</td>\n",
1032 " </tr>\n",
1033 " <tr>\n",
1034 " <td>2020-03-20</td>\n",
1035 " <td>4144</td>\n",
1036 " <td>149</td>\n",
1037 " <td>23173</td>\n",
1038 " <td>500</td>\n",
1039 " <td>593.0</td>\n",
1040 " <td>41.0</td>\n",
1041 " </tr>\n",
1042 " <tr>\n",
1043 " <td>2020-03-27</td>\n",
1044 " <td>12291</td>\n",
1045 " <td>720</td>\n",
1046 " <td>78855</td>\n",
1047 " <td>3169</td>\n",
1048 " <td>1693.0</td>\n",
1049 " <td>140.0</td>\n",
1050 " </tr>\n",
1051 " <tr>\n",
1052 " <td>2020-04-03</td>\n",
1053 " <td>25664</td>\n",
1054 " <td>2870</td>\n",
1055 " <td>217112</td>\n",
1056 " <td>15602</td>\n",
1057 " <td>2221.0</td>\n",
1058 " <td>469.0</td>\n",
1059 " </tr>\n",
1060 " <tr>\n",
1061 " <td>2020-04-10</td>\n",
1062 " <td>33254</td>\n",
1063 " <td>5868</td>\n",
1064 " <td>433554</td>\n",
1065 " <td>47578</td>\n",
1066 " <td>218.0</td>\n",
1067 " <td>458.0</td>\n",
1068 " </tr>\n",
1069 " <tr>\n",
1070 " <td>2020-04-17</td>\n",
1071 " <td>29808</td>\n",
1072 " <td>6340</td>\n",
1073 " <td>654431</td>\n",
1074 " <td>92676</td>\n",
1075 " <td>-66.0</td>\n",
1076 " <td>-81.0</td>\n",
1077 " </tr>\n",
1078 " <tr>\n",
1079 " <td>2020-04-24</td>\n",
1080 " <td>33923</td>\n",
1081 " <td>5845</td>\n",
1082 " <td>880467</td>\n",
1083 " <td>135580</td>\n",
1084 " <td>422.0</td>\n",
1085 " <td>-302.0</td>\n",
1086 " </tr>\n",
1087 " <tr>\n",
1088 " <td>2020-05-01</td>\n",
1089 " <td>32226</td>\n",
1090 " <td>4987</td>\n",
1091 " <td>1110138</td>\n",
1092 " <td>173156</td>\n",
1093 " <td>-45.0</td>\n",
1094 " <td>-53.0</td>\n",
1095 " </tr>\n",
1096 " <tr>\n",
1097 " <td>2020-05-08</td>\n",
1098 " <td>26812</td>\n",
1099 " <td>3850</td>\n",
1100 " <td>1320759</td>\n",
1101 " <td>203075</td>\n",
1102 " <td>-1615.0</td>\n",
1103 " <td>-134.0</td>\n",
1104 " </tr>\n",
1105 " <tr>\n",
1106 " <td>2020-05-15</td>\n",
1107 " <td>21611</td>\n",
1108 " <td>3006</td>\n",
1109 " <td>1481545</td>\n",
1110 " <td>226425</td>\n",
1111 " <td>-520.0</td>\n",
1112 " <td>-112.0</td>\n",
1113 " </tr>\n",
1114 " <tr>\n",
1115 " <td>2020-05-22</td>\n",
1116 " <td>17430</td>\n",
1117 " <td>2449</td>\n",
1118 " <td>1614993</td>\n",
1119 " <td>245286</td>\n",
1120 " <td>-589.0</td>\n",
1121 " <td>-90.0</td>\n",
1122 " </tr>\n",
1123 " <tr>\n",
1124 " <td>2020-05-29</td>\n",
1125 " <td>12658</td>\n",
1126 " <td>2199</td>\n",
1127 " <td>1722647</td>\n",
1128 " <td>261258</td>\n",
1129 " <td>-883.0</td>\n",
1130 " <td>78.0</td>\n",
1131 " </tr>\n",
1132 " <tr>\n",
1133 " <td>2020-06-05</td>\n",
1134 " <td>9772</td>\n",
1135 " <td>1697</td>\n",
1136 " <td>1797791</td>\n",
1137 " <td>274939</td>\n",
1138 " <td>-479.0</td>\n",
1139 " <td>-239.0</td>\n",
1140 " </tr>\n",
1141 " <tr>\n",
1142 " <td>2020-06-12</td>\n",
1143 " <td>7341</td>\n",
1144 " <td>1388</td>\n",
1145 " <td>1855247</td>\n",
1146 " <td>285795</td>\n",
1147 " <td>-157.0</td>\n",
1148 " <td>-25.0</td>\n",
1149 " </tr>\n",
1150 " <tr>\n",
1151 " <td>2020-06-19</td>\n",
1152 " <td>6939</td>\n",
1153 " <td>1018</td>\n",
1154 " <td>1905027</td>\n",
1155 " <td>293710</td>\n",
1156 " <td>-186.0</td>\n",
1157 " <td>-15.0</td>\n",
1158 " </tr>\n",
1159 " <tr>\n",
1160 " <td>2020-06-26</td>\n",
1161 " <td>5899</td>\n",
1162 " <td>835</td>\n",
1163 " <td>1950419</td>\n",
1164 " <td>300001</td>\n",
1165 " <td>-235.0</td>\n",
1166 " <td>12.0</td>\n",
1167 " </tr>\n",
1168 " <tr>\n",
1169 " <td>2020-07-03</td>\n",
1170 " <td>4485</td>\n",
1171 " <td>765</td>\n",
1172 " <td>1985555</td>\n",
1173 " <td>305684</td>\n",
1174 " <td>-127.0</td>\n",
1175 " <td>-60.0</td>\n",
1176 " </tr>\n",
1177 " <tr>\n",
1178 " <td>2020-07-10</td>\n",
1179 " <td>4131</td>\n",
1180 " <td>607</td>\n",
1181 " <td>2014685</td>\n",
1182 " <td>310295</td>\n",
1183 " <td>42.0</td>\n",
1184 " <td>-4.0</td>\n",
1185 " </tr>\n",
1186 " <tr>\n",
1187 " <td>2020-07-17</td>\n",
1188 " <td>4266</td>\n",
1189 " <td>517</td>\n",
1190 " <td>2044059</td>\n",
1191 " <td>314237</td>\n",
1192 " <td>79.0</td>\n",
1193 " <td>-19.0</td>\n",
1194 " </tr>\n",
1195 " <tr>\n",
1196 " <td>2020-07-24</td>\n",
1197 " <td>4496</td>\n",
1198 " <td>435</td>\n",
1199 " <td>2074665</td>\n",
1200 " <td>317595</td>\n",
1201 " <td>1.0</td>\n",
1202 " <td>-13.0</td>\n",
1203 " </tr>\n",
1204 " <tr>\n",
1205 " <td>2020-07-31</td>\n",
1206 " <td>3869</td>\n",
1207 " <td>445</td>\n",
1208 " <td>2104314</td>\n",
1209 " <td>320764</td>\n",
1210 " <td>73.0</td>\n",
1211 " <td>-15.0</td>\n",
1212 " </tr>\n",
1213 " <tr>\n",
1214 " <td>2020-08-07</td>\n",
1215 " <td>5833</td>\n",
1216 " <td>414</td>\n",
1217 " <td>2139062</td>\n",
1218 " <td>323799</td>\n",
1219 " <td>104.0</td>\n",
1220 " <td>11.0</td>\n",
1221 " </tr>\n",
1222 " <tr>\n",
1223 " <td>2020-08-14</td>\n",
1224 " <td>4655</td>\n",
1225 " <td>113</td>\n",
1226 " <td>1554023</td>\n",
1227 " <td>232703</td>\n",
1228 " <td>198.0</td>\n",
1229 " <td>-49.0</td>\n",
1230 " </tr>\n",
1231 " </tbody>\n",
1232 "</table>\n",
1233 "</div>"
1234 ],
1235 "text/plain": [
1236 " cases deaths cases_culm deaths_culm cases_diff deaths_diff\n",
1237 "dateRep \n",
1238 "2020-01-03 0 0 0 0 0.0 0.0\n",
1239 "2020-01-10 0 0 0 0 0.0 0.0\n",
1240 "2020-01-17 0 0 0 0 0.0 0.0\n",
1241 "2020-01-24 0 0 0 0 0.0 0.0\n",
1242 "2020-01-31 0 0 0 0 0.0 0.0\n",
1243 "2020-02-07 4 0 18 0 1.0 0.0\n",
1244 "2020-02-14 6 0 54 0 -1.0 0.0\n",
1245 "2020-02-21 0 0 70 0 0.0 0.0\n",
1246 "2020-02-28 12 0 96 0 4.0 0.0\n",
1247 "2020-03-06 198 0 681 0 52.0 0.0\n",
1248 "2020-03-13 1062 9 4279 31 350.0 2.0\n",
1249 "2020-03-20 4144 149 23173 500 593.0 41.0\n",
1250 "2020-03-27 12291 720 78855 3169 1693.0 140.0\n",
1251 "2020-04-03 25664 2870 217112 15602 2221.0 469.0\n",
1252 "2020-04-10 33254 5868 433554 47578 218.0 458.0\n",
1253 "2020-04-17 29808 6340 654431 92676 -66.0 -81.0\n",
1254 "2020-04-24 33923 5845 880467 135580 422.0 -302.0\n",
1255 "2020-05-01 32226 4987 1110138 173156 -45.0 -53.0\n",
1256 "2020-05-08 26812 3850 1320759 203075 -1615.0 -134.0\n",
1257 "2020-05-15 21611 3006 1481545 226425 -520.0 -112.0\n",
1258 "2020-05-22 17430 2449 1614993 245286 -589.0 -90.0\n",
1259 "2020-05-29 12658 2199 1722647 261258 -883.0 78.0\n",
1260 "2020-06-05 9772 1697 1797791 274939 -479.0 -239.0\n",
1261 "2020-06-12 7341 1388 1855247 285795 -157.0 -25.0\n",
1262 "2020-06-19 6939 1018 1905027 293710 -186.0 -15.0\n",
1263 "2020-06-26 5899 835 1950419 300001 -235.0 12.0\n",
1264 "2020-07-03 4485 765 1985555 305684 -127.0 -60.0\n",
1265 "2020-07-10 4131 607 2014685 310295 42.0 -4.0\n",
1266 "2020-07-17 4266 517 2044059 314237 79.0 -19.0\n",
1267 "2020-07-24 4496 435 2074665 317595 1.0 -13.0\n",
1268 "2020-07-31 3869 445 2104314 320764 73.0 -15.0\n",
1269 "2020-08-07 5833 414 2139062 323799 104.0 11.0\n",
1270 "2020-08-14 4655 113 1554023 232703 198.0 -49.0"
1271 ]
1272 },
1273 "execution_count": 20,
1274 "metadata": {},
1275 "output_type": "execute_result"
1276 }
1277 ],
1278 "source": [
1279 "data_by_week = data_by_day.resample(pd.offsets.Week(weekday=4)).sum()\n",
1280 "data_by_week"
1281 ]
1282 },
1283 {
1284 "cell_type": "code",
1285 "execution_count": 21,
1286 "metadata": {},
1287 "outputs": [
1288 {
1289 "data": {
1290 "text/html": [
1291 "<div>\n",
1292 "<style scoped>\n",
1293 " .dataframe tbody tr th:only-of-type {\n",
1294 " vertical-align: middle;\n",
1295 " }\n",
1296 "\n",
1297 " .dataframe tbody tr th {\n",
1298 " vertical-align: top;\n",
1299 " }\n",
1300 "\n",
1301 " .dataframe thead th {\n",
1302 " text-align: right;\n",
1303 " }\n",
1304 "</style>\n",
1305 "<table border=\"1\" class=\"dataframe\">\n",
1306 " <thead>\n",
1307 " <tr style=\"text-align: right;\">\n",
1308 " <th></th>\n",
1309 " <th>total_2020</th>\n",
1310 " <th>previous_mean</th>\n",
1311 " <th>covid_deaths</th>\n",
1312 " </tr>\n",
1313 " </thead>\n",
1314 " <tbody>\n",
1315 " <tr>\n",
1316 " <td>2020-03-20</td>\n",
1317 " <td>12112.0</td>\n",
1318 " <td>12007.4</td>\n",
1319 " <td>149</td>\n",
1320 " </tr>\n",
1321 " <tr>\n",
1322 " <td>2020-03-27</td>\n",
1323 " <td>12507.0</td>\n",
1324 " <td>11549.6</td>\n",
1325 " <td>720</td>\n",
1326 " </tr>\n",
1327 " <tr>\n",
1328 " <td>2020-04-03</td>\n",
1329 " <td>18565.0</td>\n",
1330 " <td>11681.4</td>\n",
1331 " <td>2870</td>\n",
1332 " </tr>\n",
1333 " <tr>\n",
1334 " <td>2020-04-10</td>\n",
1335 " <td>20929.0</td>\n",
1336 " <td>11919.4</td>\n",
1337 " <td>5868</td>\n",
1338 " </tr>\n",
1339 " <tr>\n",
1340 " <td>2020-04-17</td>\n",
1341 " <td>24691.0</td>\n",
1342 " <td>11850.6</td>\n",
1343 " <td>6340</td>\n",
1344 " </tr>\n",
1345 " <tr>\n",
1346 " <td>2020-04-24</td>\n",
1347 " <td>24303.0</td>\n",
1348 " <td>11844.4</td>\n",
1349 " <td>5845</td>\n",
1350 " </tr>\n",
1351 " <tr>\n",
1352 " <td>2020-05-01</td>\n",
1353 " <td>20059.0</td>\n",
1354 " <td>11318.4</td>\n",
1355 " <td>4987</td>\n",
1356 " </tr>\n",
1357 " <tr>\n",
1358 " <td>2020-05-08</td>\n",
1359 " <td>14428.0</td>\n",
1360 " <td>10887.2</td>\n",
1361 " <td>3850</td>\n",
1362 " </tr>\n",
1363 " <tr>\n",
1364 " <td>2020-05-15</td>\n",
1365 " <td>16390.0</td>\n",
1366 " <td>11547.0</td>\n",
1367 " <td>3006</td>\n",
1368 " </tr>\n",
1369 " <tr>\n",
1370 " <td>2020-05-22</td>\n",
1371 " <td>13839.0</td>\n",
1372 " <td>11281.0</td>\n",
1373 " <td>2449</td>\n",
1374 " </tr>\n",
1375 " <tr>\n",
1376 " <td>2020-05-29</td>\n",
1377 " <td>11265.0</td>\n",
1378 " <td>9448.0</td>\n",
1379 " <td>2199</td>\n",
1380 " </tr>\n",
1381 " <tr>\n",
1382 " <td>2020-06-05</td>\n",
1383 " <td>12106.0</td>\n",
1384 " <td>11325.6</td>\n",
1385 " <td>1697</td>\n",
1386 " </tr>\n",
1387 " <tr>\n",
1388 " <td>2020-06-12</td>\n",
1389 " <td>11302.0</td>\n",
1390 " <td>10703.6</td>\n",
1391 " <td>1388</td>\n",
1392 " </tr>\n",
1393 " <tr>\n",
1394 " <td>2020-06-19</td>\n",
1395 " <td>10694.0</td>\n",
1396 " <td>10698.2</td>\n",
1397 " <td>1018</td>\n",
1398 " </tr>\n",
1399 " <tr>\n",
1400 " <td>2020-06-26</td>\n",
1401 " <td>10282.0</td>\n",
1402 " <td>10605.6</td>\n",
1403 " <td>835</td>\n",
1404 " </tr>\n",
1405 " <tr>\n",
1406 " <td>2020-07-03</td>\n",
1407 " <td>10412.0</td>\n",
1408 " <td>10483.0</td>\n",
1409 " <td>765</td>\n",
1410 " </tr>\n",
1411 " <tr>\n",
1412 " <td>2020-07-10</td>\n",
1413 " <td>9941.0</td>\n",
1414 " <td>10509.4</td>\n",
1415 " <td>607</td>\n",
1416 " </tr>\n",
1417 " <tr>\n",
1418 " <td>2020-07-17</td>\n",
1419 " <td>10096.0</td>\n",
1420 " <td>10360.6</td>\n",
1421 " <td>517</td>\n",
1422 " </tr>\n",
1423 " <tr>\n",
1424 " <td>2020-07-24</td>\n",
1425 " <td>10159.0</td>\n",
1426 " <td>10311.6</td>\n",
1427 " <td>435</td>\n",
1428 " </tr>\n",
1429 " <tr>\n",
1430 " <td>2020-07-31</td>\n",
1431 " <td>10262.0</td>\n",
1432 " <td>10307.4</td>\n",
1433 " <td>445</td>\n",
1434 " </tr>\n",
1435 " </tbody>\n",
1436 "</table>\n",
1437 "</div>"
1438 ],
1439 "text/plain": [
1440 " total_2020 previous_mean covid_deaths\n",
1441 "2020-03-20 12112.0 12007.4 149\n",
1442 "2020-03-27 12507.0 11549.6 720\n",
1443 "2020-04-03 18565.0 11681.4 2870\n",
1444 "2020-04-10 20929.0 11919.4 5868\n",
1445 "2020-04-17 24691.0 11850.6 6340\n",
1446 "2020-04-24 24303.0 11844.4 5845\n",
1447 "2020-05-01 20059.0 11318.4 4987\n",
1448 "2020-05-08 14428.0 10887.2 3850\n",
1449 "2020-05-15 16390.0 11547.0 3006\n",
1450 "2020-05-22 13839.0 11281.0 2449\n",
1451 "2020-05-29 11265.0 9448.0 2199\n",
1452 "2020-06-05 12106.0 11325.6 1697\n",
1453 "2020-06-12 11302.0 10703.6 1388\n",
1454 "2020-06-19 10694.0 10698.2 1018\n",
1455 "2020-06-26 10282.0 10605.6 835\n",
1456 "2020-07-03 10412.0 10483.0 765\n",
1457 "2020-07-10 9941.0 10509.4 607\n",
1458 "2020-07-17 10096.0 10360.6 517\n",
1459 "2020-07-24 10159.0 10311.6 435\n",
1460 "2020-07-31 10262.0 10307.4 445"
1461 ]
1462 },
1463 "execution_count": 21,
1464 "metadata": {},
1465 "output_type": "execute_result"
1466 }
1467 ],
1468 "source": [
1469 "excess_deaths = deaths_by_week.loc['2020-03-20':, ['total_2020', 'previous_mean']].merge(\n",
1470 " data_by_week['deaths'], left_index=True, right_index=True)\n",
1471 "excess_deaths.rename(columns={'deaths': 'covid_deaths'}, inplace=True)\n",
1472 "excess_deaths.dropna(inplace=True)\n",
1473 "excess_deaths"
1474 ]
1475 },
1476 {
1477 "cell_type": "code",
1478 "execution_count": 22,
1479 "metadata": {},
1480 "outputs": [
1481 {
1482 "data": {
1483 "text/html": [
1484 "<div>\n",
1485 "<style scoped>\n",
1486 " .dataframe tbody tr th:only-of-type {\n",
1487 " vertical-align: middle;\n",
1488 " }\n",
1489 "\n",
1490 " .dataframe tbody tr th {\n",
1491 " vertical-align: top;\n",
1492 " }\n",
1493 "\n",
1494 " .dataframe thead th {\n",
1495 " text-align: right;\n",
1496 " }\n",
1497 "</style>\n",
1498 "<table border=\"1\" class=\"dataframe\">\n",
1499 " <thead>\n",
1500 " <tr style=\"text-align: right;\">\n",
1501 " <th></th>\n",
1502 " <th>total_2020</th>\n",
1503 " <th>previous_mean</th>\n",
1504 " <th>covid_deaths</th>\n",
1505 " <th>excess</th>\n",
1506 " </tr>\n",
1507 " </thead>\n",
1508 " <tbody>\n",
1509 " <tr>\n",
1510 " <td>2020-03-20</td>\n",
1511 " <td>12112.0</td>\n",
1512 " <td>12007.4</td>\n",
1513 " <td>149</td>\n",
1514 " <td>104.6</td>\n",
1515 " </tr>\n",
1516 " <tr>\n",
1517 " <td>2020-03-27</td>\n",
1518 " <td>12507.0</td>\n",
1519 " <td>11549.6</td>\n",
1520 " <td>720</td>\n",
1521 " <td>957.4</td>\n",
1522 " </tr>\n",
1523 " <tr>\n",
1524 " <td>2020-04-03</td>\n",
1525 " <td>18565.0</td>\n",
1526 " <td>11681.4</td>\n",
1527 " <td>2870</td>\n",
1528 " <td>6883.6</td>\n",
1529 " </tr>\n",
1530 " <tr>\n",
1531 " <td>2020-04-10</td>\n",
1532 " <td>20929.0</td>\n",
1533 " <td>11919.4</td>\n",
1534 " <td>5868</td>\n",
1535 " <td>9009.6</td>\n",
1536 " </tr>\n",
1537 " <tr>\n",
1538 " <td>2020-04-17</td>\n",
1539 " <td>24691.0</td>\n",
1540 " <td>11850.6</td>\n",
1541 " <td>6340</td>\n",
1542 " <td>12840.4</td>\n",
1543 " </tr>\n",
1544 " <tr>\n",
1545 " <td>2020-04-24</td>\n",
1546 " <td>24303.0</td>\n",
1547 " <td>11844.4</td>\n",
1548 " <td>5845</td>\n",
1549 " <td>12458.6</td>\n",
1550 " </tr>\n",
1551 " <tr>\n",
1552 " <td>2020-05-01</td>\n",
1553 " <td>20059.0</td>\n",
1554 " <td>11318.4</td>\n",
1555 " <td>4987</td>\n",
1556 " <td>8740.6</td>\n",
1557 " </tr>\n",
1558 " <tr>\n",
1559 " <td>2020-05-08</td>\n",
1560 " <td>14428.0</td>\n",
1561 " <td>10887.2</td>\n",
1562 " <td>3850</td>\n",
1563 " <td>3540.8</td>\n",
1564 " </tr>\n",
1565 " <tr>\n",
1566 " <td>2020-05-15</td>\n",
1567 " <td>16390.0</td>\n",
1568 " <td>11547.0</td>\n",
1569 " <td>3006</td>\n",
1570 " <td>4843.0</td>\n",
1571 " </tr>\n",
1572 " <tr>\n",
1573 " <td>2020-05-22</td>\n",
1574 " <td>13839.0</td>\n",
1575 " <td>11281.0</td>\n",
1576 " <td>2449</td>\n",
1577 " <td>2558.0</td>\n",
1578 " </tr>\n",
1579 " <tr>\n",
1580 " <td>2020-05-29</td>\n",
1581 " <td>11265.0</td>\n",
1582 " <td>9448.0</td>\n",
1583 " <td>2199</td>\n",
1584 " <td>1817.0</td>\n",
1585 " </tr>\n",
1586 " <tr>\n",
1587 " <td>2020-06-05</td>\n",
1588 " <td>12106.0</td>\n",
1589 " <td>11325.6</td>\n",
1590 " <td>1697</td>\n",
1591 " <td>780.4</td>\n",
1592 " </tr>\n",
1593 " <tr>\n",
1594 " <td>2020-06-12</td>\n",
1595 " <td>11302.0</td>\n",
1596 " <td>10703.6</td>\n",
1597 " <td>1388</td>\n",
1598 " <td>598.4</td>\n",
1599 " </tr>\n",
1600 " <tr>\n",
1601 " <td>2020-06-19</td>\n",
1602 " <td>10694.0</td>\n",
1603 " <td>10698.2</td>\n",
1604 " <td>1018</td>\n",
1605 " <td>-4.2</td>\n",
1606 " </tr>\n",
1607 " <tr>\n",
1608 " <td>2020-06-26</td>\n",
1609 " <td>10282.0</td>\n",
1610 " <td>10605.6</td>\n",
1611 " <td>835</td>\n",
1612 " <td>-323.6</td>\n",
1613 " </tr>\n",
1614 " <tr>\n",
1615 " <td>2020-07-03</td>\n",
1616 " <td>10412.0</td>\n",
1617 " <td>10483.0</td>\n",
1618 " <td>765</td>\n",
1619 " <td>-71.0</td>\n",
1620 " </tr>\n",
1621 " <tr>\n",
1622 " <td>2020-07-10</td>\n",
1623 " <td>9941.0</td>\n",
1624 " <td>10509.4</td>\n",
1625 " <td>607</td>\n",
1626 " <td>-568.4</td>\n",
1627 " </tr>\n",
1628 " <tr>\n",
1629 " <td>2020-07-17</td>\n",
1630 " <td>10096.0</td>\n",
1631 " <td>10360.6</td>\n",
1632 " <td>517</td>\n",
1633 " <td>-264.6</td>\n",
1634 " </tr>\n",
1635 " <tr>\n",
1636 " <td>2020-07-24</td>\n",
1637 " <td>10159.0</td>\n",
1638 " <td>10311.6</td>\n",
1639 " <td>435</td>\n",
1640 " <td>-152.6</td>\n",
1641 " </tr>\n",
1642 " <tr>\n",
1643 " <td>2020-07-31</td>\n",
1644 " <td>10262.0</td>\n",
1645 " <td>10307.4</td>\n",
1646 " <td>445</td>\n",
1647 " <td>-45.4</td>\n",
1648 " </tr>\n",
1649 " </tbody>\n",
1650 "</table>\n",
1651 "</div>"
1652 ],
1653 "text/plain": [
1654 " total_2020 previous_mean covid_deaths excess\n",
1655 "2020-03-20 12112.0 12007.4 149 104.6\n",
1656 "2020-03-27 12507.0 11549.6 720 957.4\n",
1657 "2020-04-03 18565.0 11681.4 2870 6883.6\n",
1658 "2020-04-10 20929.0 11919.4 5868 9009.6\n",
1659 "2020-04-17 24691.0 11850.6 6340 12840.4\n",
1660 "2020-04-24 24303.0 11844.4 5845 12458.6\n",
1661 "2020-05-01 20059.0 11318.4 4987 8740.6\n",
1662 "2020-05-08 14428.0 10887.2 3850 3540.8\n",
1663 "2020-05-15 16390.0 11547.0 3006 4843.0\n",
1664 "2020-05-22 13839.0 11281.0 2449 2558.0\n",
1665 "2020-05-29 11265.0 9448.0 2199 1817.0\n",
1666 "2020-06-05 12106.0 11325.6 1697 780.4\n",
1667 "2020-06-12 11302.0 10703.6 1388 598.4\n",
1668 "2020-06-19 10694.0 10698.2 1018 -4.2\n",
1669 "2020-06-26 10282.0 10605.6 835 -323.6\n",
1670 "2020-07-03 10412.0 10483.0 765 -71.0\n",
1671 "2020-07-10 9941.0 10509.4 607 -568.4\n",
1672 "2020-07-17 10096.0 10360.6 517 -264.6\n",
1673 "2020-07-24 10159.0 10311.6 435 -152.6\n",
1674 "2020-07-31 10262.0 10307.4 445 -45.4"
1675 ]
1676 },
1677 "execution_count": 22,
1678 "metadata": {},
1679 "output_type": "execute_result"
1680 }
1681 ],
1682 "source": [
1683 "excess_deaths['excess'] = excess_deaths.total_2020 - excess_deaths.previous_mean\n",
1684 "excess_deaths"
1685 ]
1686 },
1687 {
1688 "cell_type": "code",
1689 "execution_count": 23,
1690 "metadata": {},
1691 "outputs": [
1692 {
1693 "data": {
1694 "text/plain": [
1695 "<matplotlib.axes._subplots.AxesSubplot at 0x7f249a3ad790>"
1696 ]
1697 },
1698 "execution_count": 23,
1699 "metadata": {},
1700 "output_type": "execute_result"
1701 },
1702 {
1703 "data": {
1704 "image/png": "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\n",
1705 "text/plain": [
1706 "<Figure size 432x288 with 1 Axes>"
1707 ]
1708 },
1709 "metadata": {
1710 "needs_background": "light"
1711 },
1712 "output_type": "display_data"
1713 }
1714 ],
1715 "source": [
1716 "excess_deaths[['covid_deaths', 'excess']].plot()"
1717 ]
1718 },
1719 {
1720 "cell_type": "code",
1721 "execution_count": 24,
1722 "metadata": {},
1723 "outputs": [
1724 {
1725 "data": {
1726 "text/plain": [
1727 "1.385140247879974"
1728 ]
1729 },
1730 "execution_count": 24,
1731 "metadata": {},
1732 "output_type": "execute_result"
1733 }
1734 ],
1735 "source": [
1736 "excess_deaths.excess.sum() / excess_deaths.covid_deaths.sum()"
1737 ]
1738 },
1739 {
1740 "cell_type": "code",
1741 "execution_count": 25,
1742 "metadata": {},
1743 "outputs": [
1744 {
1745 "data": {
1746 "text/html": [
1747 "<div>\n",
1748 "<style scoped>\n",
1749 " .dataframe tbody tr th:only-of-type {\n",
1750 " vertical-align: middle;\n",
1751 " }\n",
1752 "\n",
1753 " .dataframe tbody tr th {\n",
1754 " vertical-align: top;\n",
1755 " }\n",
1756 "\n",
1757 " .dataframe thead th {\n",
1758 " text-align: right;\n",
1759 " }\n",
1760 "</style>\n",
1761 "<table border=\"1\" class=\"dataframe\">\n",
1762 " <thead>\n",
1763 " <tr style=\"text-align: right;\">\n",
1764 " <th></th>\n",
1765 " <th>total_2020</th>\n",
1766 " <th>previous_mean</th>\n",
1767 " <th>covid_deaths</th>\n",
1768 " <th>excess</th>\n",
1769 " <th>accounted_fraction</th>\n",
1770 " </tr>\n",
1771 " </thead>\n",
1772 " <tbody>\n",
1773 " <tr>\n",
1774 " <td>2020-03-20</td>\n",
1775 " <td>12112.0</td>\n",
1776 " <td>12007.4</td>\n",
1777 " <td>149</td>\n",
1778 " <td>104.6</td>\n",
1779 " <td>1.424474</td>\n",
1780 " </tr>\n",
1781 " <tr>\n",
1782 " <td>2020-03-27</td>\n",
1783 " <td>12507.0</td>\n",
1784 " <td>11549.6</td>\n",
1785 " <td>720</td>\n",
1786 " <td>957.4</td>\n",
1787 " <td>0.752037</td>\n",
1788 " </tr>\n",
1789 " <tr>\n",
1790 " <td>2020-04-03</td>\n",
1791 " <td>18565.0</td>\n",
1792 " <td>11681.4</td>\n",
1793 " <td>2870</td>\n",
1794 " <td>6883.6</td>\n",
1795 " <td>0.416933</td>\n",
1796 " </tr>\n",
1797 " <tr>\n",
1798 " <td>2020-04-10</td>\n",
1799 " <td>20929.0</td>\n",
1800 " <td>11919.4</td>\n",
1801 " <td>5868</td>\n",
1802 " <td>9009.6</td>\n",
1803 " <td>0.651305</td>\n",
1804 " </tr>\n",
1805 " <tr>\n",
1806 " <td>2020-04-17</td>\n",
1807 " <td>24691.0</td>\n",
1808 " <td>11850.6</td>\n",
1809 " <td>6340</td>\n",
1810 " <td>12840.4</td>\n",
1811 " <td>0.493754</td>\n",
1812 " </tr>\n",
1813 " <tr>\n",
1814 " <td>2020-04-24</td>\n",
1815 " <td>24303.0</td>\n",
1816 " <td>11844.4</td>\n",
1817 " <td>5845</td>\n",
1818 " <td>12458.6</td>\n",
1819 " <td>0.469154</td>\n",
1820 " </tr>\n",
1821 " <tr>\n",
1822 " <td>2020-05-01</td>\n",
1823 " <td>20059.0</td>\n",
1824 " <td>11318.4</td>\n",
1825 " <td>4987</td>\n",
1826 " <td>8740.6</td>\n",
1827 " <td>0.570556</td>\n",
1828 " </tr>\n",
1829 " <tr>\n",
1830 " <td>2020-05-08</td>\n",
1831 " <td>14428.0</td>\n",
1832 " <td>10887.2</td>\n",
1833 " <td>3850</td>\n",
1834 " <td>3540.8</td>\n",
1835 " <td>1.087325</td>\n",
1836 " </tr>\n",
1837 " <tr>\n",
1838 " <td>2020-05-15</td>\n",
1839 " <td>16390.0</td>\n",
1840 " <td>11547.0</td>\n",
1841 " <td>3006</td>\n",
1842 " <td>4843.0</td>\n",
1843 " <td>0.620690</td>\n",
1844 " </tr>\n",
1845 " <tr>\n",
1846 " <td>2020-05-22</td>\n",
1847 " <td>13839.0</td>\n",
1848 " <td>11281.0</td>\n",
1849 " <td>2449</td>\n",
1850 " <td>2558.0</td>\n",
1851 " <td>0.957389</td>\n",
1852 " </tr>\n",
1853 " <tr>\n",
1854 " <td>2020-05-29</td>\n",
1855 " <td>11265.0</td>\n",
1856 " <td>9448.0</td>\n",
1857 " <td>2199</td>\n",
1858 " <td>1817.0</td>\n",
1859 " <td>1.210237</td>\n",
1860 " </tr>\n",
1861 " <tr>\n",
1862 " <td>2020-06-05</td>\n",
1863 " <td>12106.0</td>\n",
1864 " <td>11325.6</td>\n",
1865 " <td>1697</td>\n",
1866 " <td>780.4</td>\n",
1867 " <td>2.174526</td>\n",
1868 " </tr>\n",
1869 " <tr>\n",
1870 " <td>2020-06-12</td>\n",
1871 " <td>11302.0</td>\n",
1872 " <td>10703.6</td>\n",
1873 " <td>1388</td>\n",
1874 " <td>598.4</td>\n",
1875 " <td>2.319519</td>\n",
1876 " </tr>\n",
1877 " <tr>\n",
1878 " <td>2020-06-19</td>\n",
1879 " <td>10694.0</td>\n",
1880 " <td>10698.2</td>\n",
1881 " <td>1018</td>\n",
1882 " <td>-4.2</td>\n",
1883 " <td>-242.380952</td>\n",
1884 " </tr>\n",
1885 " <tr>\n",
1886 " <td>2020-06-26</td>\n",
1887 " <td>10282.0</td>\n",
1888 " <td>10605.6</td>\n",
1889 " <td>835</td>\n",
1890 " <td>-323.6</td>\n",
1891 " <td>-2.580346</td>\n",
1892 " </tr>\n",
1893 " <tr>\n",
1894 " <td>2020-07-03</td>\n",
1895 " <td>10412.0</td>\n",
1896 " <td>10483.0</td>\n",
1897 " <td>765</td>\n",
1898 " <td>-71.0</td>\n",
1899 " <td>-10.774648</td>\n",
1900 " </tr>\n",
1901 " <tr>\n",
1902 " <td>2020-07-10</td>\n",
1903 " <td>9941.0</td>\n",
1904 " <td>10509.4</td>\n",
1905 " <td>607</td>\n",
1906 " <td>-568.4</td>\n",
1907 " <td>-1.067910</td>\n",
1908 " </tr>\n",
1909 " <tr>\n",
1910 " <td>2020-07-17</td>\n",
1911 " <td>10096.0</td>\n",
1912 " <td>10360.6</td>\n",
1913 " <td>517</td>\n",
1914 " <td>-264.6</td>\n",
1915 " <td>-1.953893</td>\n",
1916 " </tr>\n",
1917 " <tr>\n",
1918 " <td>2020-07-24</td>\n",
1919 " <td>10159.0</td>\n",
1920 " <td>10311.6</td>\n",
1921 " <td>435</td>\n",
1922 " <td>-152.6</td>\n",
1923 " <td>-2.850590</td>\n",
1924 " </tr>\n",
1925 " <tr>\n",
1926 " <td>2020-07-31</td>\n",
1927 " <td>10262.0</td>\n",
1928 " <td>10307.4</td>\n",
1929 " <td>445</td>\n",
1930 " <td>-45.4</td>\n",
1931 " <td>-9.801762</td>\n",
1932 " </tr>\n",
1933 " </tbody>\n",
1934 "</table>\n",
1935 "</div>"
1936 ],
1937 "text/plain": [
1938 " total_2020 previous_mean covid_deaths excess \\\n",
1939 "2020-03-20 12112.0 12007.4 149 104.6 \n",
1940 "2020-03-27 12507.0 11549.6 720 957.4 \n",
1941 "2020-04-03 18565.0 11681.4 2870 6883.6 \n",
1942 "2020-04-10 20929.0 11919.4 5868 9009.6 \n",
1943 "2020-04-17 24691.0 11850.6 6340 12840.4 \n",
1944 "2020-04-24 24303.0 11844.4 5845 12458.6 \n",
1945 "2020-05-01 20059.0 11318.4 4987 8740.6 \n",
1946 "2020-05-08 14428.0 10887.2 3850 3540.8 \n",
1947 "2020-05-15 16390.0 11547.0 3006 4843.0 \n",
1948 "2020-05-22 13839.0 11281.0 2449 2558.0 \n",
1949 "2020-05-29 11265.0 9448.0 2199 1817.0 \n",
1950 "2020-06-05 12106.0 11325.6 1697 780.4 \n",
1951 "2020-06-12 11302.0 10703.6 1388 598.4 \n",
1952 "2020-06-19 10694.0 10698.2 1018 -4.2 \n",
1953 "2020-06-26 10282.0 10605.6 835 -323.6 \n",
1954 "2020-07-03 10412.0 10483.0 765 -71.0 \n",
1955 "2020-07-10 9941.0 10509.4 607 -568.4 \n",
1956 "2020-07-17 10096.0 10360.6 517 -264.6 \n",
1957 "2020-07-24 10159.0 10311.6 435 -152.6 \n",
1958 "2020-07-31 10262.0 10307.4 445 -45.4 \n",
1959 "\n",
1960 " accounted_fraction \n",
1961 "2020-03-20 1.424474 \n",
1962 "2020-03-27 0.752037 \n",
1963 "2020-04-03 0.416933 \n",
1964 "2020-04-10 0.651305 \n",
1965 "2020-04-17 0.493754 \n",
1966 "2020-04-24 0.469154 \n",
1967 "2020-05-01 0.570556 \n",
1968 "2020-05-08 1.087325 \n",
1969 "2020-05-15 0.620690 \n",
1970 "2020-05-22 0.957389 \n",
1971 "2020-05-29 1.210237 \n",
1972 "2020-06-05 2.174526 \n",
1973 "2020-06-12 2.319519 \n",
1974 "2020-06-19 -242.380952 \n",
1975 "2020-06-26 -2.580346 \n",
1976 "2020-07-03 -10.774648 \n",
1977 "2020-07-10 -1.067910 \n",
1978 "2020-07-17 -1.953893 \n",
1979 "2020-07-24 -2.850590 \n",
1980 "2020-07-31 -9.801762 "
1981 ]
1982 },
1983 "execution_count": 25,
1984 "metadata": {},
1985 "output_type": "execute_result"
1986 }
1987 ],
1988 "source": [
1989 "excess_deaths['accounted_fraction'] = excess_deaths.covid_deaths / excess_deaths.excess\n",
1990 "excess_deaths"
1991 ]
1992 },
1993 {
1994 "cell_type": "code",
1995 "execution_count": 26,
1996 "metadata": {},
1997 "outputs": [
1998 {
1999 "data": {
2000 "text/html": [
2001 "<div>\n",
2002 "<style scoped>\n",
2003 " .dataframe tbody tr th:only-of-type {\n",
2004 " vertical-align: middle;\n",
2005 " }\n",
2006 "\n",
2007 " .dataframe tbody tr th {\n",
2008 " vertical-align: top;\n",
2009 " }\n",
2010 "\n",
2011 " .dataframe thead th {\n",
2012 " text-align: right;\n",
2013 " }\n",
2014 "</style>\n",
2015 "<table border=\"1\" class=\"dataframe\">\n",
2016 " <thead>\n",
2017 " <tr style=\"text-align: right;\">\n",
2018 " <th></th>\n",
2019 " <th>total_2020</th>\n",
2020 " <th>previous_mean</th>\n",
2021 " <th>covid_deaths</th>\n",
2022 " <th>excess</th>\n",
2023 " <th>accounted_fraction</th>\n",
2024 " <th>covid_deaths_m2</th>\n",
2025 " <th>excess_m2</th>\n",
2026 " <th>accounted_fraction_m2</th>\n",
2027 " </tr>\n",
2028 " </thead>\n",
2029 " <tbody>\n",
2030 " <tr>\n",
2031 " <td>2020-03-20</td>\n",
2032 " <td>12112.0</td>\n",
2033 " <td>12007.4</td>\n",
2034 " <td>149</td>\n",
2035 " <td>104.6</td>\n",
2036 " <td>1.424474</td>\n",
2037 " <td>149.0</td>\n",
2038 " <td>104.6</td>\n",
2039 " <td>1.424474</td>\n",
2040 " </tr>\n",
2041 " <tr>\n",
2042 " <td>2020-03-27</td>\n",
2043 " <td>12507.0</td>\n",
2044 " <td>11549.6</td>\n",
2045 " <td>720</td>\n",
2046 " <td>957.4</td>\n",
2047 " <td>0.752037</td>\n",
2048 " <td>434.5</td>\n",
2049 " <td>531.0</td>\n",
2050 " <td>0.818267</td>\n",
2051 " </tr>\n",
2052 " <tr>\n",
2053 " <td>2020-04-03</td>\n",
2054 " <td>18565.0</td>\n",
2055 " <td>11681.4</td>\n",
2056 " <td>2870</td>\n",
2057 " <td>6883.6</td>\n",
2058 " <td>0.416933</td>\n",
2059 " <td>1795.0</td>\n",
2060 " <td>3920.5</td>\n",
2061 " <td>0.457850</td>\n",
2062 " </tr>\n",
2063 " <tr>\n",
2064 " <td>2020-04-10</td>\n",
2065 " <td>20929.0</td>\n",
2066 " <td>11919.4</td>\n",
2067 " <td>5868</td>\n",
2068 " <td>9009.6</td>\n",
2069 " <td>0.651305</td>\n",
2070 " <td>4369.0</td>\n",
2071 " <td>7946.6</td>\n",
2072 " <td>0.549795</td>\n",
2073 " </tr>\n",
2074 " <tr>\n",
2075 " <td>2020-04-17</td>\n",
2076 " <td>24691.0</td>\n",
2077 " <td>11850.6</td>\n",
2078 " <td>6340</td>\n",
2079 " <td>12840.4</td>\n",
2080 " <td>0.493754</td>\n",
2081 " <td>6104.0</td>\n",
2082 " <td>10925.0</td>\n",
2083 " <td>0.558719</td>\n",
2084 " </tr>\n",
2085 " <tr>\n",
2086 " <td>2020-04-24</td>\n",
2087 " <td>24303.0</td>\n",
2088 " <td>11844.4</td>\n",
2089 " <td>5845</td>\n",
2090 " <td>12458.6</td>\n",
2091 " <td>0.469154</td>\n",
2092 " <td>6092.5</td>\n",
2093 " <td>12649.5</td>\n",
2094 " <td>0.481640</td>\n",
2095 " </tr>\n",
2096 " <tr>\n",
2097 " <td>2020-05-01</td>\n",
2098 " <td>20059.0</td>\n",
2099 " <td>11318.4</td>\n",
2100 " <td>4987</td>\n",
2101 " <td>8740.6</td>\n",
2102 " <td>0.570556</td>\n",
2103 " <td>5416.0</td>\n",
2104 " <td>10599.6</td>\n",
2105 " <td>0.510963</td>\n",
2106 " </tr>\n",
2107 " <tr>\n",
2108 " <td>2020-05-08</td>\n",
2109 " <td>14428.0</td>\n",
2110 " <td>10887.2</td>\n",
2111 " <td>3850</td>\n",
2112 " <td>3540.8</td>\n",
2113 " <td>1.087325</td>\n",
2114 " <td>4418.5</td>\n",
2115 " <td>6140.7</td>\n",
2116 " <td>0.719543</td>\n",
2117 " </tr>\n",
2118 " <tr>\n",
2119 " <td>2020-05-15</td>\n",
2120 " <td>16390.0</td>\n",
2121 " <td>11547.0</td>\n",
2122 " <td>3006</td>\n",
2123 " <td>4843.0</td>\n",
2124 " <td>0.620690</td>\n",
2125 " <td>3428.0</td>\n",
2126 " <td>4191.9</td>\n",
2127 " <td>0.817768</td>\n",
2128 " </tr>\n",
2129 " <tr>\n",
2130 " <td>2020-05-22</td>\n",
2131 " <td>13839.0</td>\n",
2132 " <td>11281.0</td>\n",
2133 " <td>2449</td>\n",
2134 " <td>2558.0</td>\n",
2135 " <td>0.957389</td>\n",
2136 " <td>2727.5</td>\n",
2137 " <td>3700.5</td>\n",
2138 " <td>0.737063</td>\n",
2139 " </tr>\n",
2140 " <tr>\n",
2141 " <td>2020-05-29</td>\n",
2142 " <td>11265.0</td>\n",
2143 " <td>9448.0</td>\n",
2144 " <td>2199</td>\n",
2145 " <td>1817.0</td>\n",
2146 " <td>1.210237</td>\n",
2147 " <td>2324.0</td>\n",
2148 " <td>2187.5</td>\n",
2149 " <td>1.062400</td>\n",
2150 " </tr>\n",
2151 " <tr>\n",
2152 " <td>2020-06-05</td>\n",
2153 " <td>12106.0</td>\n",
2154 " <td>11325.6</td>\n",
2155 " <td>1697</td>\n",
2156 " <td>780.4</td>\n",
2157 " <td>2.174526</td>\n",
2158 " <td>1948.0</td>\n",
2159 " <td>1298.7</td>\n",
2160 " <td>1.499961</td>\n",
2161 " </tr>\n",
2162 " <tr>\n",
2163 " <td>2020-06-12</td>\n",
2164 " <td>11302.0</td>\n",
2165 " <td>10703.6</td>\n",
2166 " <td>1388</td>\n",
2167 " <td>598.4</td>\n",
2168 " <td>2.319519</td>\n",
2169 " <td>1542.5</td>\n",
2170 " <td>689.4</td>\n",
2171 " <td>2.237453</td>\n",
2172 " </tr>\n",
2173 " <tr>\n",
2174 " <td>2020-06-19</td>\n",
2175 " <td>10694.0</td>\n",
2176 " <td>10698.2</td>\n",
2177 " <td>1018</td>\n",
2178 " <td>-4.2</td>\n",
2179 " <td>-242.380952</td>\n",
2180 " <td>1203.0</td>\n",
2181 " <td>297.1</td>\n",
2182 " <td>4.049142</td>\n",
2183 " </tr>\n",
2184 " <tr>\n",
2185 " <td>2020-06-26</td>\n",
2186 " <td>10282.0</td>\n",
2187 " <td>10605.6</td>\n",
2188 " <td>835</td>\n",
2189 " <td>-323.6</td>\n",
2190 " <td>-2.580346</td>\n",
2191 " <td>926.5</td>\n",
2192 " <td>-163.9</td>\n",
2193 " <td>-5.652837</td>\n",
2194 " </tr>\n",
2195 " <tr>\n",
2196 " <td>2020-07-03</td>\n",
2197 " <td>10412.0</td>\n",
2198 " <td>10483.0</td>\n",
2199 " <td>765</td>\n",
2200 " <td>-71.0</td>\n",
2201 " <td>-10.774648</td>\n",
2202 " <td>800.0</td>\n",
2203 " <td>-197.3</td>\n",
2204 " <td>-4.054739</td>\n",
2205 " </tr>\n",
2206 " <tr>\n",
2207 " <td>2020-07-10</td>\n",
2208 " <td>9941.0</td>\n",
2209 " <td>10509.4</td>\n",
2210 " <td>607</td>\n",
2211 " <td>-568.4</td>\n",
2212 " <td>-1.067910</td>\n",
2213 " <td>686.0</td>\n",
2214 " <td>-319.7</td>\n",
2215 " <td>-2.145762</td>\n",
2216 " </tr>\n",
2217 " <tr>\n",
2218 " <td>2020-07-17</td>\n",
2219 " <td>10096.0</td>\n",
2220 " <td>10360.6</td>\n",
2221 " <td>517</td>\n",
2222 " <td>-264.6</td>\n",
2223 " <td>-1.953893</td>\n",
2224 " <td>562.0</td>\n",
2225 " <td>-416.5</td>\n",
2226 " <td>-1.349340</td>\n",
2227 " </tr>\n",
2228 " <tr>\n",
2229 " <td>2020-07-24</td>\n",
2230 " <td>10159.0</td>\n",
2231 " <td>10311.6</td>\n",
2232 " <td>435</td>\n",
2233 " <td>-152.6</td>\n",
2234 " <td>-2.850590</td>\n",
2235 " <td>476.0</td>\n",
2236 " <td>-208.6</td>\n",
2237 " <td>-2.281879</td>\n",
2238 " </tr>\n",
2239 " <tr>\n",
2240 " <td>2020-07-31</td>\n",
2241 " <td>10262.0</td>\n",
2242 " <td>10307.4</td>\n",
2243 " <td>445</td>\n",
2244 " <td>-45.4</td>\n",
2245 " <td>-9.801762</td>\n",
2246 " <td>440.0</td>\n",
2247 " <td>-99.0</td>\n",
2248 " <td>-4.444444</td>\n",
2249 " </tr>\n",
2250 " </tbody>\n",
2251 "</table>\n",
2252 "</div>"
2253 ],
2254 "text/plain": [
2255 " total_2020 previous_mean covid_deaths excess \\\n",
2256 "2020-03-20 12112.0 12007.4 149 104.6 \n",
2257 "2020-03-27 12507.0 11549.6 720 957.4 \n",
2258 "2020-04-03 18565.0 11681.4 2870 6883.6 \n",
2259 "2020-04-10 20929.0 11919.4 5868 9009.6 \n",
2260 "2020-04-17 24691.0 11850.6 6340 12840.4 \n",
2261 "2020-04-24 24303.0 11844.4 5845 12458.6 \n",
2262 "2020-05-01 20059.0 11318.4 4987 8740.6 \n",
2263 "2020-05-08 14428.0 10887.2 3850 3540.8 \n",
2264 "2020-05-15 16390.0 11547.0 3006 4843.0 \n",
2265 "2020-05-22 13839.0 11281.0 2449 2558.0 \n",
2266 "2020-05-29 11265.0 9448.0 2199 1817.0 \n",
2267 "2020-06-05 12106.0 11325.6 1697 780.4 \n",
2268 "2020-06-12 11302.0 10703.6 1388 598.4 \n",
2269 "2020-06-19 10694.0 10698.2 1018 -4.2 \n",
2270 "2020-06-26 10282.0 10605.6 835 -323.6 \n",
2271 "2020-07-03 10412.0 10483.0 765 -71.0 \n",
2272 "2020-07-10 9941.0 10509.4 607 -568.4 \n",
2273 "2020-07-17 10096.0 10360.6 517 -264.6 \n",
2274 "2020-07-24 10159.0 10311.6 435 -152.6 \n",
2275 "2020-07-31 10262.0 10307.4 445 -45.4 \n",
2276 "\n",
2277 " accounted_fraction covid_deaths_m2 excess_m2 \\\n",
2278 "2020-03-20 1.424474 149.0 104.6 \n",
2279 "2020-03-27 0.752037 434.5 531.0 \n",
2280 "2020-04-03 0.416933 1795.0 3920.5 \n",
2281 "2020-04-10 0.651305 4369.0 7946.6 \n",
2282 "2020-04-17 0.493754 6104.0 10925.0 \n",
2283 "2020-04-24 0.469154 6092.5 12649.5 \n",
2284 "2020-05-01 0.570556 5416.0 10599.6 \n",
2285 "2020-05-08 1.087325 4418.5 6140.7 \n",
2286 "2020-05-15 0.620690 3428.0 4191.9 \n",
2287 "2020-05-22 0.957389 2727.5 3700.5 \n",
2288 "2020-05-29 1.210237 2324.0 2187.5 \n",
2289 "2020-06-05 2.174526 1948.0 1298.7 \n",
2290 "2020-06-12 2.319519 1542.5 689.4 \n",
2291 "2020-06-19 -242.380952 1203.0 297.1 \n",
2292 "2020-06-26 -2.580346 926.5 -163.9 \n",
2293 "2020-07-03 -10.774648 800.0 -197.3 \n",
2294 "2020-07-10 -1.067910 686.0 -319.7 \n",
2295 "2020-07-17 -1.953893 562.0 -416.5 \n",
2296 "2020-07-24 -2.850590 476.0 -208.6 \n",
2297 "2020-07-31 -9.801762 440.0 -99.0 \n",
2298 "\n",
2299 " accounted_fraction_m2 \n",
2300 "2020-03-20 1.424474 \n",
2301 "2020-03-27 0.818267 \n",
2302 "2020-04-03 0.457850 \n",
2303 "2020-04-10 0.549795 \n",
2304 "2020-04-17 0.558719 \n",
2305 "2020-04-24 0.481640 \n",
2306 "2020-05-01 0.510963 \n",
2307 "2020-05-08 0.719543 \n",
2308 "2020-05-15 0.817768 \n",
2309 "2020-05-22 0.737063 \n",
2310 "2020-05-29 1.062400 \n",
2311 "2020-06-05 1.499961 \n",
2312 "2020-06-12 2.237453 \n",
2313 "2020-06-19 4.049142 \n",
2314 "2020-06-26 -5.652837 \n",
2315 "2020-07-03 -4.054739 \n",
2316 "2020-07-10 -2.145762 \n",
2317 "2020-07-17 -1.349340 \n",
2318 "2020-07-24 -2.281879 \n",
2319 "2020-07-31 -4.444444 "
2320 ]
2321 },
2322 "execution_count": 26,
2323 "metadata": {},
2324 "output_type": "execute_result"
2325 }
2326 ],
2327 "source": [
2328 "excess_deaths['covid_deaths_m2'] = excess_deaths.covid_deaths.transform(lambda x: x.rolling(2, 1).mean())\n",
2329 "excess_deaths['excess_m2'] = excess_deaths.excess.transform(lambda x: x.rolling(2, 1).mean())\n",
2330 "excess_deaths['accounted_fraction_m2'] = excess_deaths.covid_deaths_m2 / excess_deaths.excess_m2\n",
2331 "excess_deaths"
2332 ]
2333 },
2334 {
2335 "cell_type": "code",
2336 "execution_count": 27,
2337 "metadata": {},
2338 "outputs": [
2339 {
2340 "data": {
2341 "text/plain": [
2342 "<matplotlib.axes._subplots.AxesSubplot at 0x7f249a356990>"
2343 ]
2344 },
2345 "execution_count": 27,
2346 "metadata": {},
2347 "output_type": "execute_result"
2348 },
2349 {
2350 "data": {
2351 "image/png": "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\n",
2352 "text/plain": [
2353 "<Figure size 432x288 with 1 Axes>"
2354 ]
2355 },
2356 "metadata": {
2357 "needs_background": "light"
2358 },
2359 "output_type": "display_data"
2360 }
2361 ],
2362 "source": [
2363 "excess_deaths[['covid_deaths', 'excess', 'covid_deaths_m2', 'excess_m2']].plot()"
2364 ]
2365 },
2366 {
2367 "cell_type": "code",
2368 "execution_count": 28,
2369 "metadata": {},
2370 "outputs": [
2371 {
2372 "data": {
2373 "text/plain": [
2374 "<matplotlib.axes._subplots.AxesSubplot at 0x7f249a26fb50>"
2375 ]
2376 },
2377 "execution_count": 28,
2378 "metadata": {},
2379 "output_type": "execute_result"
2380 },
2381 {
2382 "data": {
2383 "image/png": "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\n",
2384 "text/plain": [
2385 "<Figure size 432x288 with 1 Axes>"
2386 ]
2387 },
2388 "metadata": {
2389 "needs_background": "light"
2390 },
2391 "output_type": "display_data"
2392 }
2393 ],
2394 "source": [
2395 "excess_deaths[['accounted_fraction', 'accounted_fraction_m2']].plot()"
2396 ]
2397 },
2398 {
2399 "cell_type": "code",
2400 "execution_count": 29,
2401 "metadata": {},
2402 "outputs": [
2403 {
2404 "data": {
2405 "text/plain": [
2406 "1397"
2407 ]
2408 },
2409 "execution_count": 29,
2410 "metadata": {},
2411 "output_type": "execute_result"
2412 }
2413 ],
2414 "source": [
2415 "excess_deaths.tail(3).covid_deaths.sum()"
2416 ]
2417 },
2418 {
2419 "cell_type": "code",
2420 "execution_count": 30,
2421 "metadata": {},
2422 "outputs": [
2423 {
2424 "data": {
2425 "text/plain": [
2426 "-0.3311381531853975"
2427 ]
2428 },
2429 "execution_count": 30,
2430 "metadata": {},
2431 "output_type": "execute_result"
2432 }
2433 ],
2434 "source": [
2435 "excess_deaths.tail(3).excess.sum() / excess_deaths.tail(3).covid_deaths.sum()"
2436 ]
2437 },
2438 {
2439 "cell_type": "code",
2440 "execution_count": 31,
2441 "metadata": {},
2442 "outputs": [
2443 {
2444 "data": {
2445 "text/plain": [
2446 "1"
2447 ]
2448 },
2449 "execution_count": 31,
2450 "metadata": {},
2451 "output_type": "execute_result"
2452 }
2453 ],
2454 "source": [
2455 "max(1, excess_deaths.tail(3).excess.sum() / excess_deaths.tail(3).covid_deaths.sum())"
2456 ]
2457 },
2458 {
2459 "cell_type": "code",
2460 "execution_count": 32,
2461 "metadata": {},
2462 "outputs": [],
2463 "source": [
2464 "with open('excess_death_accuracy.json', 'w') as f:\n",
2465 " json.dump(max(1, excess_deaths.tail(3).excess.sum() / excess_deaths.tail(3).covid_deaths.sum()), f)"
2466 ]
2467 },
2468 {
2469 "cell_type": "code",
2470 "execution_count": null,
2471 "metadata": {},
2472 "outputs": [],
2473 "source": []
2474 }
2475 ],
2476 "metadata": {
2477 "kernelspec": {
2478 "display_name": "Python 3",
2479 "language": "python",
2480 "name": "python3"
2481 },
2482 "language_info": {
2483 "codemirror_mode": {
2484 "name": "ipython",
2485 "version": 3
2486 },
2487 "file_extension": ".py",
2488 "mimetype": "text/x-python",
2489 "name": "python",
2490 "nbconvert_exporter": "python",
2491 "pygments_lexer": "ipython3",
2492 "version": "3.7.4"
2493 }
2494 },
2495 "nbformat": 4,
2496 "nbformat_minor": 4
2497 }
Code for analysing Covid-19 data