COVID-19 short-term forecasts Deaths 2020-05-09


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 The documentation that is provided is still in progress and not peer reviewed.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.

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) 2020-05-10 to 2020-05-16

DateUKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
2020-05-09 31587 115002 615 8581 210 276 7549 526 26478 265 26271 405 1446 30395 5422 785 1126 939 3220 219 1830
2020-05-10 32000 116000 620 8670 210 280 7640 530 26700 270 26400 410 1460 30600 5460 800 1140 960 3260 220 1840
2020-05-11 32500 117000 620 8780 210 280 7740 540 26900 270 26700 420 1480 30900 5510 810 1150 980 3330 220 1850
2020-05-12 33000 118000 620 8900 220 290 7850 540 27100 280 26900 430 1510 31100 5560 830 1160 1000 3410 220 1860
2020-05-13 33500 119000 630 9020 220 290 7950 550 27300 280 27100 440 1530 31400 5620 840 1180 1030 3480 220 1870
2020-05-14 34000 121000 630 9130 220 300 8060 560 27500 290 27300 450 1550 31600 5680 860 1190 1050 3560 220 1880
2020-05-15 34500 122000 630 9250 220 300 8170 560 27700 290 27500 460 1570 31900 5740 880 1200 1080 3640 230 1890
2020-05-16 35000 123000 630 9380 220 310 8280 570 27900 300 27700 470 1590 32200 5800 890 1220 1100 3730 230 1900

Deaths count average forecast (bold black line in graphs) 2020-05-10 to 2020-05-16

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesRussiaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NVUS-NYUS-OHUS-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-09 10656 4823 2101 959 6589 3353 704 1827 3739 78795 390 531 2687 962 2929 311 221 1715 1381 252 3346 1362 304 2193 4836 1513 4477 558 478 421 551 9116 301 26612 1331 270 3779 330 237 1066 827 921 398
2020-05-10 11200 4980 2220 970 6640 3590 720 1910 3790 80300 410 560 2750 980 2990 320 220 1750 1410 260 3490 1390 310 2220 4920 1540 4490 580 490 440 560 9400 310 26900 1370 270 3910 340 240 1080 850 930 400
2020-05-11 11900 5140 2340 990 6700 3840 730 2020 3840 82100 430 600 2820 1000 3070 330 230 1800 1440 270 3650 1430 320 2250 5040 1590 4540 610 510 460 580 9700 310 27200 1420 280 4070 340 240 1090 880 940 410
2020-05-12 12700 5310 2480 1010 6760 4110 740 2130 3890 83800 450 640 2890 1020 3140 340 240 1850 1480 290 3820 1470 330 2290 5170 1640 4600 640 540 480 590 10000 320 27500 1470 280 4240 350 250 1110 910 950 410
2020-05-13 13500 5490 2620 1030 6820 4400 760 2240 3950 85600 470 680 2970 1030 3210 350 250 1910 1520 300 4000 1500 330 2320 5300 1690 4660 680 560 500 610 10400 330 27800 1520 290 4410 360 250 1140 940 970 420
2020-05-14 14400 5670 2770 1050 6880 4710 780 2360 4000 87400 490 730 3040 1050 3290 360 260 1960 1550 310 4190 1540 340 2350 5440 1740 4730 710 580 520 630 10800 330 28100 1570 290 4600 370 260 1160 970 980 430
2020-05-15 15400 5860 2930 1070 6940 5030 790 2490 4060 89300 510 780 3120 1070 3370 380 270 2020 1600 320 4390 1580 350 2390 5580 1800 4790 740 610 550 650 11200 340 28500 1630 300 4790 370 260 1190 1000 1000 440
2020-05-16 16400 6060 3100 1090 7000 5390 810 2630 4110 91300 540 840 3200 1100 3450 390 280 2080 1640 340 4590 1620 360 2420 5720 1850 4860 780 640 570 670 11700 350 28800 1680 300 4990 380 270 1210 1040 1010 450

Deaths count forecast Europe (bold red line in graphs) 2020-05-10 to 2020-05-16

DateUKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
2020-05-09 31587 115002 615 8581 210 276 7549 526 26478 265 26271 405 1446 30395 5422 785 1126 939 3220 219 1830
2020-05-10 32000 116000 620 8640 210 280 7620 530 26700 270 26500 410 1470 30600 5480 800 1140 960 3290 220 1840
2020-05-11 32500 117000 620 8710 210 280 7690 540 26800 270 26600 420 1490 30900 5530 820 1150 980 3370 220 1850
2020-05-12 33000 118000 620 8780 220 290 7770 540 27000 280 26800 430 1510 31100 5590 830 1160 1000 3450 220 1850
2020-05-13 33400 119000 620 8850 220 290 7850 550 27200 280 27000 440 1530 31300 5640 850 1170 1020 3530 220 1860
2020-05-14 33900 120000 630 8920 220 290 7930 550 27400 290 27200 460 1550 31500 5700 870 1180 1050 3620 220 1870
2020-05-15 34400 121000 630 9000 230 290 8010 560 27600 290 27300 470 1580 31800 5760 880 1200 1070 3700 230 1880
2020-05-16 34900 122000 630 9070 230 300 8090 560 27800 300 27500 480 1600 32000 5820 900 1210 1090 3790 230 1890

Deaths count forecast (bold red line in graphs) 2020-05-10 to 2020-05-16

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesRussiaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NVUS-NYUS-OHUS-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-09 10656 4823 2101 959 6589 3353 704 1827 3739 78795 390 531 2687 962 2929 311 221 1715 1381 252 3346 1362 304 2193 4836 1513 4477 558 478 421 551 9116 301 26612 1331 270 3779 330 237 1066 827 921 398
2020-05-10 11500 4960 2220 980 6640 3640 710 1940 3790 80500 410 570 2760 970 2990 320 230 1770 1400 260 3520 1390 310 2220 4970 1550 4530 580 510 430 570 9300 310 26900 1380 280 3880 340 240 1100 850 940 410
2020-05-11 12400 5110 2340 990 6690 3920 720 2060 3830 82100 430 600 2830 990 3060 330 240 1820 1420 270 3690 1430 310 2260 5100 1600 4610 600 530 440 590 9600 310 27200 1420 280 3980 350 240 1130 880 950 420
2020-05-12 13400 5260 2470 1010 6740 4230 740 2180 3880 83800 450 640 2900 1000 3120 350 240 1880 1440 290 3860 1460 320 2290 5230 1650 4690 630 560 450 610 9800 320 27500 1460 280 4090 350 240 1160 900 960 430
2020-05-13 14400 5410 2610 1030 6790 4550 750 2310 3930 85500 470 680 2970 1010 3190 360 250 1940 1460 300 4040 1490 320 2320 5360 1700 4760 650 580 460 630 10100 330 27800 1510 290 4200 360 250 1190 930 980 440
2020-05-14 15500 5560 2750 1050 6840 4890 760 2450 3970 87200 490 720 3050 1030 3260 370 260 2000 1480 310 4230 1530 330 2350 5500 1750 4840 680 610 470 650 10500 340 28100 1550 290 4310 370 250 1220 960 990 450
2020-05-15 16600 5720 2900 1070 6890 5260 770 2600 4020 88900 510 760 3130 1040 3330 390 270 2060 1500 320 4430 1560 340 2380 5640 1800 4920 710 640 490 670 10800 350 28500 1600 300 4420 380 250 1250 990 1000 460
2020-05-16 17900 5890 3060 1080 6940 5660 780 2760 4070 90700 540 800 3210 1060 3400 400 280 2120 1520 340 4630 1600 340 2420 5780 1850 5000 740 670 500 690 11200 350 28800 1650 310 4540 380 260 1280 1020 1020 470

Deaths count scenario forecast (bold purple line in graphs) 2020-05-10 to 2020-05-18

DateUKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCHBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesRussiaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NVUS-NYUS-OHUS-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-09 31587 115002 615 8581 210 276 7549 526 26478 265 26271 405 1446 30395 5422 785 1126 939 3220 219 1830 10656 4823 2101 959 6589 3353 704 1827 3739 78795 390 531 2687 962 2929 311 221 1715 1381 252 3346 1362 304 2193 4836 1513 4477 558 478 421 551 9116 301 26612 1331 270 3779 330 237 1066 827 921 398
2020-05-10 32000 116000 620 8700 210 280 7660 530 26600 270 26500 410 1470 30600 5450 800 1140 960 3300 220 1840 11300 4960 2180 980 6650 3580 710 1900 3780 80200 410 570 2730 980 2980 320 230 1750 1410 260 3450 1390 310 2210 4930 1550 4500 580 500 440 560 9200 300 26900 1370 270 3940 340 240 1070 850 930 400
2020-05-11 32400 117000 620 8810 210 280 7760 530 26800 270 26700 420 1490 30800 5490 820 1150 980 3360 220 1850 12000 5050 2260 990 6700 3800 730 1970 3830 81800 430 600 2790 1000 3020 330 230 1800 1430 270 3540 1420 310 2240 5040 1590 4580 600 510 450 580 9400 310 27100 1410 280 4110 340 240 1090 860 950 410
2020-05-12 32800 118000 620 8890 220 280 7850 540 26900 280 26900 420 1510 31000 5520 830 1160 990 3430 220 1860 12600 5140 2330 1000 6750 4010 730 2040 3870 83400 440 620 2840 1010 3050 340 240 1840 1450 280 3620 1450 310 2270 5160 1640 4630 630 530 460 590 9500 310 27300 1450 280 4220 350 250 1120 880 960 420
2020-05-13 33100 119000 630 8950 220 290 7920 540 27100 280 27000 420 1530 31200 5560 840 1170 1000 3500 220 1870 13300 5230 2390 1020 6790 4210 740 2120 3910 84800 450 640 2890 1020 3080 350 250 1870 1470 290 3700 1480 320 2290 5280 1680 4670 650 540 470 610 9700 310 27600 1490 280 4360 360 250 1140 890 960 420
2020-05-14 33500 120000 630 9010 220 290 7950 540 27200 280 27100 430 1550 31300 5600 850 1180 1020 3550 220 1880 13900 5330 2440 1020 6830 4410 750 2180 3950 86100 460 640 2940 1040 3110 360 250 1900 1480 300 3770 1500 320 2310 5400 1720 4710 660 550 480 620 9800 320 27800 1530 290 4490 370 250 1160 900 970 430
2020-05-15 33700 120000 630 9060 220 290 8030 550 27400 280 27300 430 1560 31400 5630 860 1190 1030 3570 220 1880 14600 5420 2510 1030 6870 4610 760 2230 3990 87200 470 640 2980 1050 3140 360 260 1930 1500 300 3840 1530 320 2330 5510 1760 4760 680 570 480 630 9900 320 28000 1560 290 4610 380 250 1180 900 980 430
2020-05-16 34000 121000 640 9090 220 290 8080 550 27500 290 27300 440 1570 31600 5670 870 1190 1040 3600 220 1890 15200 5510 2560 1040 6890 4800 760 2290 4020 88100 480 680 3020 1060 3170 370 260 1960 1520 310 3900 1560 320 2340 5600 1790 4810 690 580 490 640 10000 320 28100 1600 290 4710 390 250 1180 920 990 440
2020-05-17 34300 121000 640 9130 220 290 8120 550 27600 290 27400 440 1580 31700 5700 880 1200 1060 3640 220 1900 15900 5600 2620 1050 6910 4980 760 2350 4050 88900 490 690 3060 1070 3190 380 270 1990 1540 310 3950 1580 330 2360 5680 1830 4870 700 580 500 640 10200 320 28300 1630 290 4810 390 250 1200 930 990 440
2020-05-18 34500 122000 640 9140 220 300 8190 560 27700 290 27500 440 1590 31800 5740 890 1210 1070 3670 230 1910 16500 5670 2680 1060 6930 5150 770 2410 4080 89700 500 690 3090 1070 3220 380 270 2020 1560 320 4020 1610 330 2380 5730 1870 4910 710 590 500 650 10300 330 28400 1670 300 4910 400 250 1220 940 990 450

Peak increase in estimated trend of Deaths in Europe 2020-05-09

UKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
Peak date04-0904-0704-0804-1004-1104-0904-1504-0404-0204-2104-0704-1804-2403-2804-0704-2404-1105-0104-2204-0804-04
Peak daily increment 1014 3241 23 364 8 11 280 18 902 19 1014 14 112 826 163 30 33 26 117 9 61
Days from 100 to peak 22 34 9 17 1 0 23 3 20 1 23 5 22 24 18 18 13 29 25 -1 12
Days from peak/2 to peak 16 24 17 15 22 14 22 14 16 23 17 25 24 19 18 28 19 37 27 21 17
Days since peak 30 32 31 29 28 30 24 35 37 18 32 21 15 42 32 15 28 8 17 31 35

Peak increase in estimated trend of Deaths 2020-05-09

BrazilCanadaIndiaIndonesiaIranMexicoPhilippinesRussiaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NVUS-NYUS-OHUS-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
Peak date --05-0405-0404-1403-19 --04-12 --04-1904-15 -- --04-2304-2404-2004-2504-28 --04-07 -- --04-2904-1604-1504-2504-2904-22 -- -- -- --04-2904-1204-0704-2904-21 --04-2904-0304-3004-2204-0604-05
Peak daily increment 175 119 39 150 28 125 2279 89 58 120 13 9 46 50 13 75 271 63 148 304 11 1066 58 10 20 10 37 58 32 13
Days from 100 to peak 33 29 16 14 11 22 29 26 20 18 5 4 7 25 1 19 24 22 25 28 0 18 25 8 14 -8 26 7 14 -3
Days from peak/2 to peak 34 32 26 16 25 24 21 30 27 20 26 29 11 32 20 22 23 27 26 28 17 13 31 26 33 5 35 13 29 11
Days since peak 5 5 25 51 27 20 24 16 15 19 14 11 32 10 23 24 14 10 17 10 27 32 10 18 10 36 9 17 33 34

Initial visual evaluation of forecasts of Deaths