COVID-19 short-term forecasts Deaths 2020-05-18


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 The documentation that is provided is still in progress and not peer reviewed.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.
[2020-04-16] Added scenario forecasts to all graphs now. This would now be the preferred forecast for most.
This is the first time with a peak in confirmed UK cases (also for deaths, but this is uncertain because it is at the same date).
[2020-04-17] Bird and Nielsen look into nowcasting death counts in England.
[2020-04-24] A summary of our work on short-term COVID-19 forecasting appeared as a voxeu.
[2020-04-27] Our short-term COVID-19 forecasting paper is now available as Nuffield Economics Discussion Paper 2020-W06.
A small adjustment has been made to the scenario forecast methodology, and will be documented shortly.
[2020-04-29] See our blog entry at the International Institute of Forecasters.
US history of death counts revised in Johns Hopkins/CSSE data.
UK death counts have been revised to include the deaths in care homes. In the Johns Hopkins/CSSE data set, which we use, the entire history has been revised. So forecasts made up to 2020-04-29 cannot be compared to later outcomes. In the ECDC data set only the last observation has changed, causing a jump in the series.
[2020-05-06] The New York Times is in the process of redefining its US state data. Unfortunately, at the moment only the last observation has changed (e.g New York deaths jumped from 19645 on 2020-05-05 to 25956 a day later). This means the data is currently useless; however it does bring it close to the Johns Hopkins/CSSE count (25626 on 2020-05-06). The aggregate US count is based on JH/CSSE so unaffected. We now use Johns Hopkins/CSSE US state data, including all states with sufficient counts. So the new forecasts cannot be compared to those previously.
A minor change is that we show the graph without scenario forecast if no peak has been detected yet.
[2020-05-13] We now omit countries with fewer than 200 confirmed cases in the last week (25 for deaths).
The short-term paper has some small updates, including further comparisons with other models.
Data for Ecuador are not reliable enough for forecasting.
Switched to an improved version of scenario forecasting.
[2020-05-18] Minor fixes to the improved version of scenario forecasting, backported to 2020-05-13.

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. Also available as Nuffield Economics Discussion Paper 2020-W06. Still preliminary is the documentation of the medium term forecasts.

Deaths count average forecast Europe (bold black line in graphs) 2020-05-19 to 2020-05-25

DateUKEUBEDEDKESFIFRHUIEITNLPLPTROSE
2020-05-18 34796 122228 9080 8003 548 27709 300 28193 462 1547 32007 5694 936 1231 1120 3698
2020-05-19 35100 123000 9120 8040 550 27800 300 28300 460 1560 32200 5720 950 1240 1140 3730
2020-05-20 35400 124000 9170 8070 550 27900 310 28500 470 1570 32300 5750 960 1240 1150 3780
2020-05-21 35700 124000 9220 8110 560 28100 310 28600 470 1580 32500 5780 980 1260 1170 3840
2020-05-22 36100 125000 9270 8140 560 28200 310 28800 480 1590 32700 5810 990 1270 1190 3890
2020-05-23 36400 126000 9320 8180 560 28300 320 29000 480 1600 32900 5840 1010 1280 1210 3950
2020-05-24 36700 127000 9370 8210 560 28400 320 29200 490 1610 33100 5870 1030 1290 1230 4010
2020-05-25 37100 128000 9420 8250 570 28600 330 29400 490 1620 33200 5910 1040 1300 1250 4070

Deaths count average forecast (bold black line in graphs) 2020-05-19 to 2020-05-25

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-18 16853 5960 478 592 3156 1191 7057 5332 2789 831 2722 286 4171 90347 489 686 3279 1223 3448 392 296 1997 1624 355 4233 1621 346 2440 5858 1937 4855 731 591 527 693 10439 270 353 28339 1657 4515 391 296 1350 1015 1002 459
2020-05-19 17500 6050 500 610 3280 1200 7090 5530 2910 840 2810 290 4210 91300 490 700 3330 1250 3500 400 300 2020 1640 360 4300 1640 350 2460 5950 1970 4890 750 600 540 700 10600 280 360 28500 1680 4560 400 300 1370 1030 1010 460
2020-05-20 18300 6160 530 620 3430 1210 7140 5790 3060 850 2910 310 4250 92500 500 720 3390 1280 3560 410 310 2050 1670 370 4400 1670 350 2490 6040 2020 4930 770 610 550 720 10800 280 360 28700 1710 4640 400 310 1410 1050 1020 470
2020-05-21 19100 6270 560 640 3590 1230 7180 6060 3210 870 3020 320 4300 93700 510 740 3460 1310 3610 420 330 2090 1700 390 4490 1700 350 2510 6140 2060 4970 780 620 560 730 11100 290 370 28900 1750 4720 410 310 1440 1070 1020 480
2020-05-22 20000 6380 590 650 3750 1250 7230 6340 3370 880 3130 330 4340 94900 520 750 3520 1340 3670 430 340 2120 1730 400 4590 1720 360 2540 6250 2110 5010 800 640 570 750 11400 300 380 29000 1780 4800 410 320 1470 1090 1030 480
2020-05-23 20900 6490 620 670 3920 1270 7280 6640 3540 890 3240 340 4380 96100 530 770 3590 1370 3730 440 350 2150 1760 410 4700 1750 360 2560 6350 2160 5050 820 650 580 760 11700 310 380 29200 1820 4890 420 320 1510 1110 1040 490
2020-05-24 21800 6600 660 690 4100 1300 7330 6960 3720 910 3360 360 4430 97400 540 790 3660 1410 3780 450 360 2190 1790 420 4800 1780 360 2590 6460 2210 5090 840 660 590 780 12000 320 390 29400 1860 4980 420 330 1540 1130 1050 500
2020-05-25 22800 6720 690 710 4290 1320 7380 7290 3910 920 3490 370 4470 98700 550 820 3730 1440 3840 470 370 2220 1820 440 4910 1810 370 2610 6570 2260 5130 860 670 600 800 12300 330 400 29600 1890 5070 430 330 1580 1150 1060 500

Deaths count forecast Europe (bold red line in graphs) 2020-05-19 to 2020-05-25

DateUKEUBEDEDKESFIFRHUIEITNLPLPTROSE
2020-05-18 34796 122228 9080 8003 548 27709 300 28193 462 1547 32007 5694 936 1231 1120 3698
2020-05-19 35100 123000 9120 8040 550 27800 300 28400 470 1560 32200 5720 950 1240 1140 3740
2020-05-20 35400 124000 9160 8070 550 27900 310 28600 470 1570 32300 5740 960 1250 1150 3780
2020-05-21 35700 124000 9210 8110 560 28000 310 28800 480 1570 32500 5770 970 1260 1160 3810
2020-05-22 36000 125000 9250 8140 560 28100 310 29000 490 1580 32600 5790 990 1270 1180 3850
2020-05-23 36300 126000 9290 8180 560 28300 310 29200 490 1590 32800 5820 1000 1280 1190 3890
2020-05-24 36600 126000 9330 8210 560 28400 320 29500 500 1600 32900 5840 1010 1300 1210 3930
2020-05-25 36900 127000 9370 8240 570 28500 320 29700 500 1610 33100 5860 1020 1310 1220 3970

Deaths count forecast (bold red line in graphs) 2020-05-19 to 2020-05-25

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-18 16853 5960 478 592 3156 1191 7057 5332 2789 831 2722 286 4171 90347 489 686 3279 1223 3448 392 296 1997 1624 355 4233 1621 346 2440 5858 1937 4855 731 591 527 693 10439 270 353 28339 1657 4515 391 296 1350 1015 1002 459
2020-05-19 17600 6040 510 610 3280 1220 7110 5510 2930 840 2810 300 4210 91300 490 700 3330 1240 3490 400 300 2020 1650 360 4300 1640 350 2460 5930 1960 4890 740 600 540 710 10600 280 360 28500 1690 4570 400 300 1370 1030 1010 460
2020-05-20 18300 6120 540 620 3420 1250 7160 5700 3080 850 2910 310 4250 92400 500 710 3380 1260 3530 410 310 2050 1690 370 4370 1670 350 2480 6000 1990 4930 760 610 540 720 10800 280 360 28700 1710 4620 400 300 1380 1040 1010 470
2020-05-21 19100 6200 570 640 3560 1280 7220 5880 3230 860 3000 330 4280 93300 500 720 3430 1270 3580 420 320 2070 1720 380 4440 1690 360 2510 6070 2020 4960 770 620 550 730 11000 290 370 28800 1740 4680 400 310 1400 1060 1020 480
2020-05-22 19900 6280 600 650 3710 1310 7270 6080 3390 870 3100 340 4320 94400 510 730 3480 1290 3620 420 330 2100 1760 390 4520 1710 360 2530 6140 2050 4990 780 630 560 740 11200 300 370 29000 1770 4740 410 310 1420 1070 1030 480
2020-05-23 20700 6350 630 670 3870 1340 7320 6280 3560 880 3200 360 4360 95400 510 740 3540 1310 3660 430 340 2120 1790 400 4590 1730 370 2550 6210 2080 5020 800 640 570 750 11500 300 380 29100 1800 4790 410 320 1440 1090 1030 490
2020-05-24 21600 6440 660 680 4030 1370 7370 6480 3730 890 3310 370 4390 96400 520 760 3590 1330 3710 440 350 2150 1840 400 4660 1750 370 2570 6290 2110 5050 810 650 570 760 11700 310 380 29300 1830 4860 420 320 1450 1100 1040 490
2020-05-25 22400 6520 700 700 4200 1410 7430 6700 3920 900 3420 390 4430 97500 520 770 3640 1350 3750 450 360 2170 1880 410 4740 1780 380 2590 6360 2150 5080 830 650 580 770 12000 320 390 29400 1860 4920 420 330 1470 1110 1050 500

Deaths count scenario forecast (bold purple line in graphs) 2020-05-19 to 2020-05-27

DateUKEUBEDEDKESFIFRHUIEITNLPLPTROSEBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-18 34796 122228 9080 8003 548 27709 300 28193 462 1547 32007 5694 936 1231 1120 3698 16853 5960 478 592 3156 1191 7057 5332 2789 831 2722 286 4171 90347 489 686 3279 1223 3448 392 296 1997 1624 355 4233 1621 346 2440 5858 1937 4855 731 591 527 693 10439 270 353 28339 1657 4515 391 296 1350 1015 1002 459
2020-05-19 35100 123000 9130 8040 550 27800 300 28400 460 1560 32200 5730 950 1240 1140 3770 17500 6070 510 610 3280 1220 7090 5570 2920 840 2850 290 4210 91400 500 710 3330 1250 3500 400 310 2030 1640 370 4330 1650 350 2460 5960 1980 4890 750 600 540 710 10600 280 360 28500 1690 4630 390 300 1410 1040 1010 460
2020-05-20 35400 124000 9180 8080 550 27900 310 28500 470 1570 32300 5750 960 1250 1150 3810 18200 6170 540 620 3420 1260 7120 5800 3060 850 2970 310 4240 92300 510 730 3370 1280 3560 410 330 2060 1660 380 4420 1670 350 2480 6050 2020 4920 770 610 550 720 10700 290 360 28700 1720 4690 400 300 1450 1050 1010 460
2020-05-21 35600 124000 9220 8120 560 28000 310 28700 470 1580 32500 5770 970 1260 1170 3850 18900 6260 570 640 3540 1290 7160 6020 3200 850 3100 320 4270 93100 520 740 3410 1310 3610 410 340 2080 1670 390 4550 1690 360 2490 6140 2070 4950 780 620 560 740 10800 300 370 28800 1750 4750 400 310 1490 1070 1020 470
2020-05-22 35800 125000 9250 8160 560 28100 310 28800 480 1580 32600 5800 980 1260 1180 3890 19500 6350 600 650 3670 1330 7190 6240 3340 850 3210 330 4290 93900 530 760 3450 1330 3660 420 350 2110 1690 400 4650 1710 360 2510 6230 2100 4970 800 620 560 750 10900 310 370 29000 1780 4810 400 310 1550 1090 1020 470
2020-05-23 36000 125000 9290 8190 560 28100 310 28900 480 1590 32800 5820 990 1270 1190 3940 20100 6420 630 670 3780 1360 7220 6440 3510 860 3310 340 4310 94600 540 770 3490 1340 3680 430 360 2130 1700 410 4740 1730 360 2530 6300 2140 5000 810 630 570 760 11000 320 370 29100 1810 4870 410 310 1590 1100 1020 470
2020-05-24 36200 126000 9320 8230 560 28200 310 29100 490 1600 32900 5840 990 1280 1200 3970 20700 6490 660 680 3910 1400 7250 6640 3660 860 3420 350 4340 95400 540 780 3520 1350 3720 430 370 2150 1710 420 4810 1750 370 2540 6370 2180 5020 820 640 570 770 11000 330 380 29300 1830 4930 410 320 1630 1120 1030 470
2020-05-25 36400 126000 9350 8260 570 28300 320 29200 490 1610 33000 5880 1000 1280 1210 4000 21200 6560 690 700 4130 1420 7270 6820 3820 870 3520 370 4360 96000 550 800 3560 1350 3750 440 380 2170 1720 420 4870 1760 370 2550 6440 2210 5050 840 640 580 780 11100 340 380 29400 1860 4980 410 320 1670 1130 1030 480
2020-05-26 36600 127000 9390 8280 570 28300 320 29300 500 1620 33100 5900 1010 1290 1220 4040 21700 6620 710 710 4210 1450 7300 7000 3960 870 3620 380 4380 96800 560 800 3590 1370 3770 440 380 2190 1730 430 4910 1780 370 2560 6500 2250 5080 850 640 580 780 11200 350 380 29500 1880 5030 420 320 1710 1140 1040 480
2020-05-27 36800 127000 9420 8310 570 28400 320 29300 500 1620 33200 5920 1020 1290 1230 4060 22200 6690 740 720 4280 1470 7320 7170 4070 880 3710 390 4410 97500 570 810 3630 1370 3800 450 390 2210 1740 440 4960 1790 380 2570 6550 2270 5100 860 650 590 790 11300 350 390 29600 1900 5070 420 330 1740 1150 1040 480

Peak increase in estimated trend of Deaths in Europe 2020-05-18

UKEUBEDEDKESFIFRHUIEITNLPLPTROSE
Peak date04-0904-0704-1004-1504-0404-0204-2104-0704-1804-2403-2804-0704-2404-1104-2904-22
Peak daily increment 1014 3241 364 280 18 903 19 1014 14 110 826 163 29 33 26 117
Days from 100 to peak 22 34 17 23 3 20 1 23 5 22 24 18 18 13 27 25
Days from peak/2 to peak 16 24 15 22 14 16 23 17 25 25 19 18 28 19 35 27
Days since peak 39 41 38 33 44 46 27 41 30 24 51 41 24 37 19 26

Peak increase in estimated trend of Deaths 2020-05-18

BrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
Peak date --05-05 --05-05 -- --03-19 -- --04-12 -- --04-1904-1505-0605-0804-2304-2404-2004-25 --05-0704-0705-1305-1304-2904-1604-1504-2504-2904-2205-0605-0805-0605-1304-2905-1304-1204-0704-2905-0704-2904-0305-1404-2204-0604-05
Peak daily increment 178 18 150 28 125 2279 18 38 89 57 120 13 54 46 14 127 50 13 75 265 64 148 25 21 24 20 305 11 11 1066 60 175 21 10 40 58 32 13
Days from 100 to peak 34 23 14 11 22 29 23 26 26 20 18 5 36 7 19 43 25 1 19 24 22 25 19 27 23 30 28 16 0 18 25 34 14 -8 40 7 14 -3
Days from peak/2 to peak 35 37 16 25 24 21 37 36 30 27 20 26 42 11 42 44 32 20 22 23 27 26 33 41 38 43 28 41 17 13 31 35 33 5 48 13 29 11
Days since peak 13 13 60 36 29 33 12 10 25 24 28 23 11 41 5 5 19 32 33 23 19 26 12 10 12 5 19 5 36 41 19 11 19 45 4 26 42 43

Initial visual evaluation of forecasts of Deaths