COVID-19 short-term forecasts Deaths 2020-05-11


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 The documentation that is provided is still in progress and not peer reviewed.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) 2020-05-12 to 2020-05-18

DateUKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
2020-05-11 32065 116412 620 8707 212 282 7661 533 26744 271 26604 421 1467 30739 5456 811 1144 982 3256 224 1845
2020-05-12 32400 117000 620 8780 210 290 7720 540 26900 270 26700 430 1480 30900 5490 820 1150 1000 3290 220 1850
2020-05-13 32800 118000 620 8870 220 290 7800 540 27100 280 26900 440 1500 31100 5540 840 1160 1030 3360 230 1860
2020-05-14 33200 119000 630 8960 220 290 7880 550 27300 280 27100 450 1520 31400 5590 850 1170 1050 3420 230 1870
2020-05-15 33600 120000 630 9050 220 300 7960 550 27400 290 27300 460 1540 31600 5640 870 1190 1070 3490 230 1880
2020-05-16 34000 121000 630 9150 220 300 8050 560 27600 290 27500 470 1560 31800 5690 880 1200 1100 3560 230 1890
2020-05-17 34500 122000 640 9240 220 300 8130 560 27800 290 27600 480 1570 32000 5740 900 1210 1120 3630 230 1890
2020-05-18 34900 123000 640 9340 230 310 8220 570 28000 300 27800 490 1590 32300 5790 920 1220 1150 3700 230 1900

Deaths count average forecast (bold black line in graphs) 2020-05-12 to 2020-05-18

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesRussiaSouth 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-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-11 11653 5115 2294 991 6685 3573 726 2009 206 3841 80682 403 541 2779 983 3005 328 225 1735 1418 271 3457 1411 311 2240 5102 1577 4526 591 498 435 575 9340 208 305 26988 1357 274 3832 346 245 1117 850 945 409
2020-05-12 12300 5260 2400 1010 6730 3710 740 2110 220 3880 81800 410 560 2830 990 3050 340 230 1760 1440 280 3530 1440 320 2270 5220 1610 4570 610 510 450 590 9500 210 310 27200 1380 280 3890 350 250 1140 870 960 410
2020-05-13 13100 5410 2520 1020 6790 3900 750 2220 230 3930 83300 420 590 2890 1010 3110 350 240 1810 1460 290 3640 1470 320 2300 5350 1660 4630 630 530 460 610 9800 220 320 27400 1410 280 3990 350 250 1170 890 970 420
2020-05-14 14000 5570 2640 1040 6840 4090 760 2340 240 3970 84700 440 620 2950 1020 3170 360 240 1850 1500 300 3740 1500 330 2330 5490 1700 4690 660 540 480 620 10100 230 320 27700 1450 290 4100 360 250 1200 910 990 430
2020-05-15 14900 5730 2780 1060 6890 4300 770 2460 250 4020 86200 450 650 3020 1040 3230 370 250 1900 1530 320 3850 1530 330 2360 5620 1750 4750 680 560 490 640 10400 230 330 27900 1490 290 4220 370 260 1230 940 1000 440
2020-05-16 15800 5890 2920 1080 6950 4520 790 2580 260 4070 87700 470 690 3080 1050 3290 390 260 1940 1560 330 3970 1560 340 2390 5770 1800 4810 700 580 510 660 10800 240 330 28200 1520 290 4330 370 260 1260 960 1020 450
2020-05-17 16900 6070 3060 1100 7000 4750 800 2720 270 4120 89300 480 720 3150 1070 3350 400 270 1990 1600 340 4080 1600 340 2420 5910 1840 4870 730 600 520 680 11200 250 340 28400 1560 300 4460 380 260 1290 990 1040 460
2020-05-18 18000 6240 3220 1110 7060 5000 820 2860 290 4170 90900 500 760 3220 1080 3420 410 280 2040 1630 360 4210 1630 350 2450 6060 1900 4930 760 620 540 700 11600 260 350 28700 1600 300 4590 390 270 1330 1010 1050 460

Deaths count forecast Europe (bold red line in graphs) 2020-05-12 to 2020-05-18

DateUKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
2020-05-11 32065 116412 620 8707 212 282 7661 533 26744 271 26604 421 1467 30739 5456 811 1144 982 3256 224 1845
2020-05-12 32400 117000 620 8790 210 290 7740 540 26900 270 26800 430 1480 30900 5500 820 1150 1000 3300 230 1850
2020-05-13 32700 118000 620 8870 220 290 7820 540 27100 280 26900 440 1500 31100 5540 830 1160 1020 3350 230 1860
2020-05-14 33000 119000 630 8940 220 290 7890 540 27200 280 27100 450 1510 31300 5570 850 1170 1040 3390 230 1870
2020-05-15 33400 120000 630 9020 220 290 7960 550 27400 280 27300 460 1530 31500 5610 860 1180 1060 3430 230 1880
2020-05-16 33700 121000 630 9090 220 300 8040 550 27500 290 27400 460 1540 31700 5650 870 1190 1080 3480 230 1890
2020-05-17 34000 122000 630 9170 220 300 8110 550 27700 290 27600 470 1560 31900 5680 880 1200 1100 3520 230 1890
2020-05-18 34300 122000 630 9250 220 300 8190 560 27900 290 27800 480 1570 32100 5720 890 1210 1130 3570 240 1900

Deaths count forecast (bold red line in graphs) 2020-05-12 to 2020-05-18

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesRussiaSouth 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-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-11 11653 5115 2294 991 6685 3573 726 2009 206 3841 80682 403 541 2779 983 3005 328 225 1735 1418 271 3457 1411 311 2240 5102 1577 4526 591 498 435 575 9340 208 305 26988 1357 274 3832 346 245 1117 850 945 409
2020-05-12 12400 5240 2380 1010 6740 3710 740 2120 220 3880 81900 410 560 2830 990 3050 340 230 1760 1440 280 3530 1450 320 2270 5220 1610 4570 610 510 450 590 9500 220 310 27200 1380 280 3890 350 250 1150 870 960 420
2020-05-13 13100 5380 2480 1030 6780 3850 750 2240 230 3930 83100 420 580 2880 1010 3100 340 230 1790 1460 280 3610 1480 320 2300 5350 1640 4620 620 520 460 600 9700 220 310 27500 1410 280 3950 360 250 1180 880 970 420
2020-05-14 13900 5510 2580 1040 6830 4000 760 2360 240 3970 84300 430 600 2940 1020 3140 350 240 1820 1480 290 3690 1500 320 2320 5480 1670 4660 640 540 470 610 9900 230 320 27700 1430 280 4010 370 260 1210 890 980 430
2020-05-15 14800 5650 2680 1060 6880 4140 770 2480 250 4010 85500 440 620 2990 1030 3190 360 240 1850 1500 300 3770 1530 330 2350 5610 1700 4700 650 550 480 620 10100 240 320 27900 1450 290 4070 380 260 1240 910 990 440
2020-05-16 15700 5790 2780 1080 6930 4300 780 2620 270 4060 86700 450 640 3040 1040 3240 380 250 1880 1520 310 3850 1560 330 2380 5740 1730 4740 670 560 480 630 10400 250 320 28200 1480 290 4130 380 260 1270 920 1000 450
2020-05-17 16600 5930 2890 1100 6980 4460 790 2760 280 4100 87900 470 670 3100 1050 3280 390 250 1910 1540 320 3930 1600 340 2400 5880 1760 4790 690 580 490 640 10700 260 330 28400 1500 290 4180 390 270 1300 940 1010 450
2020-05-18 17700 6080 3010 1120 7030 4630 800 2910 300 4150 89200 480 690 3160 1060 3330 400 260 1950 1560 330 4010 1630 340 2430 6020 1790 4830 700 590 500 650 11000 270 330 28600 1530 300 4240 400 270 1340 950 1030 460

Deaths count scenario forecast (bold purple line in graphs) 2020-05-12 to 2020-05-20

DateUKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCHBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesRussiaSouth 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-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-11 32065 116412 620 8707 212 282 7661 533 26744 271 26604 421 1467 30739 5456 811 1144 982 3256 224 1845 11653 5115 2294 991 6685 3573 726 2009 206 3841 80682 403 541 2779 983 3005 328 225 1735 1418 271 3457 1411 311 2240 5102 1577 4526 591 498 435 575 9340 208 305 26988 1357 274 3832 346 245 1117 850 945 409
2020-05-12 32500 117000 620 8810 210 290 7730 540 26900 270 26700 430 1490 30900 5510 820 1150 1000 3310 220 1850 12400 5230 2370 1010 6730 3760 740 2080 210 3880 81800 420 570 2820 990 3040 340 230 1770 1430 280 3530 1440 310 2270 5220 1620 4600 610 520 450 590 9500 210 310 27200 1400 280 3970 350 250 1150 860 960 420
2020-05-13 32800 118000 630 8870 220 290 7790 540 27100 280 26900 440 1510 31100 5550 830 1160 1020 3350 230 1860 13000 5350 2440 1020 6770 3930 740 2160 220 3920 82800 430 580 2860 1000 3070 350 230 1790 1450 290 3600 1470 320 2290 5350 1660 4660 620 530 450 610 9700 220 320 27300 1430 280 4050 360 250 1180 870 970 430
2020-05-14 33100 119000 630 8910 220 290 7850 540 27200 280 27000 440 1530 31200 5590 840 1170 1040 3370 230 1870 13600 5460 2510 1030 6810 4100 750 2220 220 3960 83800 440 590 2890 1010 3100 350 240 1810 1470 300 3650 1500 320 2310 5450 1700 4730 630 550 460 620 9800 220 320 27300 1470 290 4110 370 250 1210 880 970 430
2020-05-15 33300 120000 630 8950 220 290 7900 550 27400 280 27100 450 1540 31400 5620 850 1180 1060 3390 230 1880 14200 5580 2560 1040 6840 4270 760 2300 230 4000 84700 450 590 2920 1020 3130 360 240 1830 1480 300 3710 1520 320 2330 5550 1750 4760 650 560 460 630 9900 230 320 27300 1490 290 4170 370 250 1230 890 980 440
2020-05-16 33600 120000 640 9000 220 290 7950 550 27500 280 27200 450 1550 31500 5660 860 1190 1080 3420 230 1880 14800 5680 2610 1060 6870 4430 770 2360 240 4030 85600 450 600 2960 1030 3160 360 240 1850 1500 310 3770 1550 320 2350 5640 1790 4840 660 560 470 640 10100 240 330 27400 1520 290 4240 380 250 1260 900 990 440
2020-05-17 33900 121000 640 9040 220 300 8000 550 27600 280 27300 460 1560 31600 5690 860 1190 1090 3440 230 1890 15300 5790 2650 1060 6900 4600 770 2390 240 4060 86500 460 610 2990 1040 3190 370 250 1870 1520 310 3830 1580 320 2370 5730 1830 4880 670 570 470 650 10200 240 330 27400 1550 290 4310 390 260 1290 910 990 450
2020-05-18 34100 121000 640 9070 220 300 8050 550 27800 290 27400 460 1570 31700 5720 870 1200 1110 3470 230 1900 15800 5880 2690 1060 6930 4750 780 2460 240 4090 87300 470 610 3020 1050 3220 380 250 1890 1540 310 3890 1600 330 2380 5800 1860 4940 690 570 480 660 10200 250 330 27400 1580 300 4420 400 260 1310 920 990 450
2020-05-19 34400 122000 640 9110 220 300 8100 560 27900 290 27500 470 1580 31900 5750 870 1210 1130 3490 230 1900 16400 5970 2740 1070 6960 4900 790 2500 250 4130 88000 470 610 3060 1060 3240 380 250 1910 1550 310 3950 1620 330 2400 5870 1890 4960 700 570 480 660 10300 250 340 27500 1610 300 4500 410 260 1330 930 1000 460
2020-05-20 34600 122000 640 9150 220 300 8150 560 28000 290 27600 470 1580 32000 5780 870 1210 1140 3520 230 1900 16800 6040 2780 1080 6980 5050 790 2530 250 4160 88900 470 610 3090 1070 3270 380 250 1940 1560 320 4000 1640 330 2420 5950 1920 4960 710 580 490 670 10400 250 340 27600 1630 300 4560 410 260 1350 940 1000 460

Peak increase in estimated trend of Deaths in Europe 2020-05-11

UKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
Peak date04-0904-0704-0804-1004-1104-0904-1504-0404-0204-2104-0704-1804-2403-2804-0704-2404-1104-2904-2204-0804-04
Peak daily increment 1014 3241 23 364 8 11 280 18 902 19 1014 14 110 826 163 29 33 25 117 9 61
Days from 100 to peak 22 34 9 17 1 0 23 3 20 1 23 5 22 24 18 18 13 27 25 -1 12
Days from peak/2 to peak 16 24 17 15 22 14 22 14 16 23 17 25 25 19 18 28 19 35 27 21 17
Days since peak 32 34 33 31 30 32 26 37 39 20 34 23 17 44 34 17 30 12 19 33 37

Peak increase in estimated trend of Deaths 2020-05-11

BrazilCanadaIndiaIndonesiaIranMexicoPhilippinesRussiaSouth 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-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
Peak date --05-0405-0404-1403-19 --04-12 -- --04-1904-1505-06 --04-2304-2404-2004-2504-28 --04-07 --05-0604-2904-1604-1504-2504-2904-2205-06 --05-06 --04-3005-0504-1204-0704-2904-2105-0504-2904-0304-3004-2204-0604-05
Peak daily increment 177 118 39 150 28 125 2279 18 89 57 120 13 9 46 123 50 13 75 265 64 148 25 23 310 8 11 1066 60 10 177 22 10 38 58 32 13
Days from 100 to peak 33 29 16 14 11 22 29 23 26 20 18 5 4 7 36 25 1 19 24 22 25 19 23 29 8 0 18 25 8 32 14 -8 26 7 14 -3
Days from peak/2 to peak 34 32 26 16 25 24 21 37 30 27 20 26 29 11 38 32 20 22 23 27 26 33 38 29 34 17 13 31 26 33 33 5 35 13 29 11
Days since peak 7 7 27 53 29 22 26 5 18 17 21 16 13 34 5 12 25 26 16 12 19 5 5 11 6 29 34 12 20 6 12 38 11 19 35 36

Initial visual evaluation of forecasts of Deaths