COVID-19 short-term forecasts Deaths 2020-04-15


Disclaimer

  • Forecasts produced by Jennie Castle, Jurgen Doornik, and David Hendry, researchers at the University of Oxford. These are our forecasts, and should not be considered official forecasts from, or endorsed by, any of: University of Oxford, Oxford Martin School, Nuffield College, or Magdalen College.
  • These forecasts are short term time-series extrapolations of the data. They are not based on epidemiological modelling or simulations. The documentation that is provided is still in progress and not peer reviewed. All forecasts are uncertain: their success can only be determined afterwards. Many mitigation strategies are in place, which, if successful, invalidate our forecasts. An explanation of our methods is provided below.

Recent changes

[2020-03-24] Our forecasts are starting to overestimate in some cases. This was always expected to happen when the increase starts to slow down. Scenario forecasts that are based on what happened in China earlier this year, but only for Italy and Spain sofar.
[2020-03-26] Scenario forecasts that are based on what happened in China earlier this year, only for Italy.
[2020-03-31] Scenario forecasts, based on what happened in China earlier this year, are presented for several countries (line marked with x). Created more plausible 90% confidence bands (dotted line in same colour).
[2020-04-02] Now including more US States, based on New York Times data. And the world.
[2020-04-06] Added a post hoc estimate of the peak number of cases. This needs at least three confirmed observations (four for deaths) after the event. It is based on the averaged smooth trend, and can change later or be a local peak. It is marked with a vertical line with the date label, or a date with left arrow in the bottom left corner of the graph. This is backported to 2020-04-04.
[2020-04-08] Minor correction to peak estimates. Added table with scenario forecasts.
[2020-04-09] Added table with estimated peak dates (if happened) and dates to and since the peak. Note that this can be a local peak, and subsequent re-acceleration (or data revisions) can result in a new peak later.
[2020-04-10] Updated documentation with better description of short-term estimates and peak determination.

Further information

  • We believe these forecasts fill a useful gap in the short run. They give an indication of what is likely to happen in the next few days, removing some aspect of surprise. Moreover, a noticeable drop in comparison to the extrapolations could be an indication that the implemented policies are having some impact. It is difficult to understand exponential growth. We hope that these forecasts may help to convince viewers to adhere to the policies implemented by their respective governments, and keep all arguments factual and measured.
  • We use the data repository for the 2019 Novel Coronavirus Visual Dashboard operated by the Johns Hopkins University Center for Systems Science and Engineering. This is updated daily, but we tend to update our forecasts only every other day.
    US state data as of 2020-03-28 is courtesy of the New York Times.
  • We can only provide forecasts of what is measured. If confirmed cases are an underestimate of actual cases, then our forecasts will also be underestimates. No other epidemiological data is used. Data definition and collection differs between countries and may change over time.
  • We will update the methodology as we learn what is happening in the next few days or weeks. Once the number of cases levels off, there is no need to provide these forecasts anymore.
  • Countries where the counts are very low or stable have been omitted.
  • The graphs have dates on the horizontal axis (yyyy-mm-dd) and cumulative counts on the vertical axis. They show
    1. bold dark grey line (with circles): observed counts (Johns Hopkins CSSE);
    2. many light grey lines (with open circles): forecasts using different model settings and starting up to four periods back;
    3. red line (with open circles): single forecasts path using default model settings;
    4. black line (with crosses): average of all forecasts, recentered on the last observation;
    5. thin green lines: some indication of uncertainty around the red forecasts, but we do not know how reliable that is.
    Both the red line forecasts and the black lines are also given in the tables above. These forecasts differ, we are currently inclined to use the average forecasts.
  • The forecasts are constructed as follows:
    1. An overall `trend' is extracted by taking a window of the data at a time. In each window we draw `straight lines' which are selected using an automatic econometric procedure (`machine learning'). All straight lines are collected and averaged, giving the trend.
    2. Forecasts are made using the estimated trend, but we note that this must be done carefully, because simply extrapolating the flexible insample trend would lead to wildly fluctuating forecast. We use the `Cardt' method, which has been found to work well in other settings.
    3. Residuals from the trend are also forecast, and combined with trend forecasts into an overall forecast.
  • Scenario forecasts are constructed very differently: smooth versions of the Chinese experience are matched at different lag lengths with the path of each country. This probably works best from the peak, or the slowdown just before (but we include it for the UK nonetheless).
  • The forecast evaluation shows past forecasts, together with the outcomes (in the grey line with circles).
  • EU-BS is Estonia, Latvia, and Lithuania together.
  • This paper describes the methodology and gives further references, but remains preliminary. Also preliminary is the documentation of the medium term forecasts.

Deaths count average forecast Europe (bold black line in graphs) 2020-04-16 to 2020-04-22

DateUKEUATBEDEDKESFRIEITNLPLPTROSECH
2020-04-15 12868 73268 393 4440 3804 309 18708 17167 444 21645 3134 286 599 372 1203 1239
2020-04-16 13700 76100 410 4760 4030 320 19200 18000 480 22200 3250 310 640 390 1320 1290
2020-04-17 14700 78800 420 5140 4220 330 19700 18700 510 22700 3360 330 670 420 1400 1330
2020-04-18 15700 81600 440 5540 4400 340 20200 19400 550 23200 3470 350 710 440 1480 1380
2020-04-19 16800 84600 450 5980 4600 350 20700 20200 590 23800 3580 370 760 460 1570 1440
2020-04-20 18000 87600 470 6460 4810 370 21200 20900 630 24300 3690 400 800 490 1670 1490
2020-04-21 19300 90800 490 6990 5030 380 21800 21800 670 24900 3810 430 850 520 1780 1550
2020-04-22 20600 94200 500 7570 5260 390 22300 22600 710 25500 3930 450 900 550 1900 1600

Deaths count average forecast (bold black line in graphs) 2020-04-16 to 2020-04-22

DateBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WA
2020-04-15 1736 1006 4777 349 28326 885 355 868 613 569 958 1103 1108 1919 3156 11586 572
2020-04-16 1920 1110 4880 380 30800 960 380 970 660 620 1040 1190 1240 2070 3470 12300 590
2020-04-17 2080 1200 4980 410 33600 1020 400 1050 700 660 1130 1260 1370 2230 3760 13100 610
2020-04-18 2260 1290 5090 450 36600 1090 430 1140 750 700 1220 1340 1510 2410 4080 13900 620
2020-04-19 2450 1400 5200 480 39900 1160 450 1250 800 750 1310 1430 1660 2600 4430 14800 640
2020-04-20 2660 1520 5310 530 43500 1240 480 1360 850 800 1420 1510 1830 2800 4810 15700 660
2020-04-21 2900 1640 5420 570 47400 1330 520 1480 910 850 1530 1610 2020 3020 5230 16700 670
2020-04-22 3150 1780 5540 620 51700 1420 550 1620 980 910 1650 1710 2230 3260 5690 17700 690

Deaths count forecast Europe (bold red line in graphs) 2020-04-16 to 2020-04-22

DateUKEUATBEDEDKESFRIEITNLPLPTROSECH
2020-04-15 12868 73268 393 4440 3804 309 18708 17167 444 21645 3134 286 599 372 1203 1239
2020-04-16 13600 75300 400 4730 3940 320 19100 17800 480 22100 3220 300 640 390 1260 1280
2020-04-17 14400 78000 410 5030 4190 330 19600 18700 520 22600 3340 320 690 410 1310 1320
2020-04-18 15100 80800 420 5330 4460 340 20100 19600 560 23100 3470 340 730 430 1360 1370
2020-04-19 15900 83600 430 5640 4740 350 20500 20500 610 23700 3600 360 770 450 1400 1420
2020-04-20 16700 86600 440 5960 5030 370 21000 21500 650 24200 3730 380 820 470 1450 1470
2020-04-21 17600 89700 450 6300 5350 380 21600 22600 710 24800 3870 400 880 490 1500 1520
2020-04-22 18500 92900 460 6650 5690 390 22100 23700 760 25300 4020 420 930 510 1560 1580

Deaths count forecast (bold red line in graphs) 2020-04-16 to 2020-04-22

DateBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WA
2020-04-15 1736 1006 4777 349 28326 885 355 868 613 569 958 1103 1108 1919 3156 11586 572
2020-04-16 1920 1090 4870 380 30700 940 370 900 630 660 1070 1190 1240 1980 3420 12200 590
2020-04-17 2110 1190 4970 410 33300 1010 390 980 660 760 1180 1290 1390 2070 3690 12900 600
2020-04-18 2330 1300 5070 430 36100 1090 410 1070 690 860 1290 1410 1550 2170 3980 13600 620
2020-04-19 2570 1410 5160 460 39200 1170 430 1170 720 980 1410 1530 1730 2270 4300 14300 630
2020-04-20 2830 1530 5260 480 42500 1260 460 1270 750 1120 1540 1650 1930 2370 4650 15000 650
2020-04-21 3120 1660 5360 510 46200 1360 480 1390 790 1270 1690 1800 2150 2490 5030 15800 670
2020-04-22 3440 1810 5460 550 50100 1470 500 1520 820 1440 1850 1950 2390 2600 5440 16600 680

Deaths count scenario forecast (bold green line in graphs) 2020-04-16 to 2020-04-24

DateUKEUATBEDEDKESFRIEITNLPLPTROSECHBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WA
2020-04-15 12868 73268 393 4440 3804 309 18708 17167 444 21645 3134 286 599 372 1203 1239 1736 1006 4777 349 28326 885 355 868 613 569 958 1103 1108 1919 3156 11586 572
2020-04-16 13600 75000 410 4740 3750 320 19100 17300 460 22100 3190 300 630 390 1170 1270 1840 1070 4880 370 29600 920 370 890 650 570 1010 1160 1210 2020 3270 12200 590
2020-04-17 14200 77400 420 5010 3900 330 19500 17900 490 22500 3290 320 650 410 1220 1310 1930 1140 4980 380 31400 970 390 970 680 590 1060 1220 1270 2130 3480 12800 610
2020-04-18 14800 79200 440 5280 4040 340 19900 18700 510 22900 3390 330 680 430 1260 1350 2050 1220 5070 400 33000 1010 410 1010 720 610 1110 1280 1350 2250 3680 13400 620
2020-04-19 15400 81300 450 5580 4180 350 20200 19500 530 23400 3490 350 710 450 1290 1380 2110 1290 5160 420 34400 1050 430 1020 750 630 1170 1340 1440 2360 3880 14000 640
2020-04-20 16000 82900 470 5840 4280 360 20400 20100 560 23800 3550 370 730 470 1330 1410 2200 1370 5240 440 35800 1080 440 1020 780 650 1230 1390 1500 2460 4070 14700 650
2020-04-21 16600 84900 480 6100 4400 360 20700 20900 580 24100 3620 380 760 480 1390 1430 2260 1430 5310 450 37200 1120 460 1050 800 670 1280 1450 1560 2570 4250 15200 660
2020-04-22 17100 86600 490 6350 4520 370 20900 21500 600 24300 3680 400 780 500 1430 1460 2360 1500 5390 470 38600 1150 480 1100 830 680 1310 1500 1620 2660 4410 15700 670
2020-04-23 17600 88100 510 6630 4620 380 21200 21900 620 24700 3740 410 800 510 1470 1480 2480 1570 5450 480 39900 1190 490 1120 860 700 1380 1550 1710 2760 4570 16200 680
2020-04-24 18100 89500 520 6880 4700 380 21400 22300 640 25000 3790 420 820 530 1500 1500 2560 1630 5500 500 41400 1220 510 1120 890 720 1430 1590 1780 2860 4720 16700 680

Peak increase in estimated trend of Deaths in Europe 2020-04-15

UKEUATBEDEDKESFRIEITNLPLPTROSECH
Peak date04-1004-0704-0804-10 --04-0404-0204-07 --03-2804-0704-08 --04-11 --04-04
Peak daily increment 868 3265 23 367 18 898 1031 826 160 20 24 61
Days from 100 to peak 22 34 9 17 3 20 23 24 18 2 9 12
Days from peak/2 to peak 16 24 17 15 14 16 16 19 18 15 17 17
Days since peak 5 8 7 5 11 13 8 18 8 7 4 11

Peak increase in estimated trend of Deaths 2020-04-15

BrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WA
Peak date -- --03-19 -- -- --04-11 -- -- -- -- -- -- -- --04-0904-06
Peak daily increment 150 24 779 30
Days from 100 to peak 14 8 18 15
Days from peak/2 to peak 16 18 14 30
Days since peak 27 4 6 9

Initial visual evaluation of forecasts of Deaths