COVID-19 short-term forecasts Deaths 2020-05-22


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 The documentation that is provided is still in progress and not peer reviewed.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.
[2020-05-20] Problem with UK confirmed cases: negative daily count. This makes the forecasts temporarily unreliable.
Updated the second paper.

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) 2020-05-23 to 2020-05-29

DateUKEUBEDEESFRHUIEITNLPLPTROSECH
2020-05-22 36393 124689 9212 8228 28628 28167 476 1592 32616 5788 982 1289 1166 3925 1903
2020-05-23 36600 125000 9250 8270 28700 28200 480 1600 32800 5810 990 1300 1180 3950 1910
2020-05-24 37000 126000 9280 8320 28800 28300 480 1610 32900 5830 1000 1310 1190 3980 1910
2020-05-25 37300 126000 9310 8360 28900 28400 490 1620 33000 5850 1010 1330 1200 4020 1910
2020-05-26 37600 127000 9350 8410 29000 28400 490 1630 33200 5870 1020 1340 1210 4050 1920
2020-05-27 37900 127000 9380 8450 29100 28500 500 1640 33300 5900 1030 1350 1220 4090 1920
2020-05-28 38200 128000 9420 8500 29200 28600 500 1650 33500 5920 1040 1360 1230 4130 1920
2020-05-29 38600 128000 9450 8540 29400 28600 510 1660 33600 5940 1050 1380 1250 4170 1930

Deaths count average forecast (bold black line in graphs) 2020-05-23 to 2020-05-29

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
2020-05-22 21048 6360 630 682 3726 1326 7300 6989 3244 857 3249 397 4276 95979 541 773 3666 1323 3635 418 321 2190 1779 424 4714 1791 391 2506 6224 2142 5094 842 653 595 775 10985 302 380 28853 1872 5010 419 311 1470 1136 1050 496
2020-05-23 22100 6400 660 700 3890 1360 7350 7200 3360 860 3370 410 4300 97000 550 780 3710 1340 3660 430 330 2210 1790 430 4790 1810 400 2520 6290 2170 5140 860 660 610 780 11100 310 380 29000 1920 5040 420 310 1490 1140 1060 500
2020-05-24 23300 6470 690 720 4070 1410 7400 7580 3490 870 3520 430 4330 98000 550 800 3780 1370 3700 430 330 2250 1810 450 4880 1840 410 2540 6360 2220 5180 880 670 620 780 11300 310 390 29100 1980 5110 430 320 1520 1160 1070 510
2020-05-25 24600 6550 740 740 4250 1460 7450 8010 3620 870 3680 460 4360 99000 560 820 3850 1390 3740 440 340 2290 1840 460 4970 1880 420 2560 6440 2260 5230 910 680 640 790 11500 320 390 29200 2050 5180 430 320 1550 1170 1080 510
2020-05-26 25900 6630 780 770 4440 1510 7510 8460 3750 880 3840 490 4390 100000 570 830 3930 1420 3780 450 350 2320 1870 480 5070 1910 420 2570 6530 2300 5270 940 700 650 800 11800 320 400 29400 2120 5260 440 320 1570 1190 1090 520
2020-05-27 27300 6710 830 790 4640 1560 7560 8930 3900 880 4020 520 4420 101000 570 850 4000 1440 3830 460 360 2360 1900 500 5170 1950 430 2590 6610 2350 5320 980 710 670 810 12000 330 400 29500 2190 5340 440 320 1600 1200 1100 530
2020-05-28 28800 6800 870 820 4850 1610 7620 9430 4050 890 4190 550 4450 102000 580 870 4080 1470 3870 470 360 2400 1940 520 5270 1980 440 2610 6690 2390 5370 1010 720 680 820 12200 330 410 29600 2260 5420 450 330 1630 1220 1110 540
2020-05-29 30300 6880 930 850 5070 1660 7680 9960 4200 890 4380 580 4470 104000 590 890 4160 1490 3920 480 370 2440 1970 540 5370 2020 450 2630 6780 2440 5410 1040 740 700 830 12500 340 420 29800 2330 5510 460 330 1660 1240 1120 540

Deaths count forecast Europe (bold red line in graphs) 2020-05-23 to 2020-05-29

DateUKEUBEDEESFRHUIEITNLPLPTROSECH
2020-05-22 36393 124689 9212 8228 28628 28167 476 1592 32616 5788 982 1289 1166 3925 1903
2020-05-23 36700 125000 9250 8270 29000 28200 480 1600 32800 5810 990 1300 1180 3970 1910
2020-05-24 37000 126000 9280 8320 29400 28300 480 1610 32900 5820 1000 1310 1180 4010 1910
2020-05-25 37300 127000 9310 8360 29800 28300 490 1620 33000 5840 1010 1330 1190 4050 1920
2020-05-26 37600 128000 9340 8400 30200 28300 490 1630 33200 5860 1020 1340 1200 4090 1920
2020-05-27 37900 128000 9370 8450 30600 28400 500 1640 33300 5880 1030 1350 1210 4130 1930
2020-05-28 38200 129000 9400 8490 31000 28400 500 1650 33400 5890 1040 1370 1220 4170 1930
2020-05-29 38500 130000 9430 8530 31400 28400 500 1660 33600 5910 1040 1380 1220 4210 1940

Deaths count forecast (bold red line in graphs) 2020-05-23 to 2020-05-29

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
2020-05-22 21048 6360 630 682 3726 1326 7300 6989 3244 857 3249 397 4276 95979 541 773 3666 1323 3635 418 321 2190 1779 424 4714 1791 391 2506 6224 2142 5094 842 653 595 775 10985 302 380 28853 1872 5010 419 311 1470 1136 1050 496
2020-05-23 22100 6430 670 710 3900 1370 7360 7400 3350 870 3410 430 4300 97000 550 790 3740 1340 3680 420 330 2230 1810 440 4810 1820 400 2520 6300 2190 5140 880 660 610 790 11100 310 390 29000 1940 5110 420 310 1490 1160 1060 500
2020-05-24 23100 6510 710 730 4070 1420 7410 7900 3450 870 3580 450 4330 98000 560 800 3810 1360 3720 430 330 2270 1840 450 4900 1850 410 2530 6370 2230 5180 910 680 620 810 11200 320 390 29100 2010 5200 430 320 1510 1180 1070 510
2020-05-25 24300 6590 750 760 4250 1470 7470 8400 3560 880 3760 480 4350 99000 570 820 3890 1370 3770 440 340 2310 1870 470 4990 1880 420 2540 6450 2270 5220 940 690 630 830 11300 320 400 29200 2080 5290 440 320 1530 1210 1070 520
2020-05-26 25400 6670 790 790 4440 1530 7520 8900 3670 890 3940 520 4370 100000 570 830 3960 1390 3810 440 340 2340 1900 480 5080 1920 430 2550 6520 2310 5270 980 700 650 850 11400 330 400 29300 2150 5380 440 320 1560 1230 1080 520
2020-05-27 26700 6740 830 820 4640 1580 7580 9500 3790 900 4140 550 4400 102000 580 850 4040 1400 3850 450 350 2380 1940 500 5170 1950 430 2570 6600 2350 5310 1020 710 660 870 11600 330 410 29400 2220 5480 450 330 1580 1250 1090 530
2020-05-28 28000 6830 880 850 4850 1640 7640 10100 3910 910 4340 590 4420 103000 590 870 4120 1420 3900 460 350 2420 1970 520 5270 1980 440 2580 6670 2400 5350 1060 730 670 890 11700 340 410 29500 2300 5570 450 330 1600 1270 1100 540
2020-05-29 29400 6910 920 880 5070 1690 7690 10700 4030 910 4550 620 4450 104000 600 880 4200 1440 3940 460 360 2460 2010 540 5360 2020 450 2590 6750 2440 5390 1100 740 690 910 11800 350 420 29600 2380 5670 460 330 1620 1300 1100 550

Deaths count scenario forecast (bold purple line in graphs) 2020-05-23 to 2020-05-31

DateUKEUBEDEESFRHUIEITNLPLPTROSECHBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth 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
2020-05-22 36393 124689 9212 8228 28628 28167 476 1592 32616 5788 982 1289 1166 3925 1903 21048 6360 630 682 3726 1326 7300 6989 3244 857 3249 397 4276 95979 541 773 3666 1323 3635 418 321 2190 1779 424 4714 1791 391 2506 6224 2142 5094 842 653 595 775 10985 302 380 28853 1872 5010 419 311 1470 1136 1050 496
2020-05-23 36700 125000 9230 8270 28600 28300 480 1600 32800 5810 990 1300 1180 3950 1900 22200 6410 670 700 3890 1370 7350 7400 3370 860 3380 430 4290 97000 540 790 3710 1340 3670 420 330 2210 1790 440 4780 1820 400 2520 6300 2170 5140 870 660 610 780 11100 310 380 29000 1950 5010 420 310 1500 1140 1060 500
2020-05-24 36900 125000 9260 8300 28600 28300 480 1610 32900 5830 1000 1310 1180 3980 1910 23300 6490 720 730 4060 1410 7400 7800 3480 860 3520 470 4310 97000 550 800 3770 1360 3710 430 330 2230 1810 460 4870 1850 410 2540 6370 2210 5180 890 670 620 780 11200 310 390 29100 2020 5060 430 310 1520 1150 1070 510
2020-05-25 37200 126000 9280 8350 28600 28400 490 1620 33000 5850 1010 1320 1190 4020 1910 24400 6540 770 750 4230 1460 7440 8200 3580 870 3670 500 4330 98000 560 810 3820 1370 3740 430 330 2250 1830 480 4960 1880 410 2550 6430 2250 5220 920 680 630 780 11200 320 390 29200 2100 5100 430 320 1540 1170 1070 510
2020-05-26 37400 126000 9310 8390 28600 28400 490 1620 33100 5870 1010 1330 1200 4050 1910 25500 6610 820 780 4420 1500 7490 8600 3680 870 3810 540 4350 99000 570 820 3870 1380 3770 440 340 2270 1850 500 5020 1900 420 2560 6490 2280 5250 950 690 640 790 11300 320 390 29400 2170 5140 440 320 1560 1180 1080 520
2020-05-27 37600 127000 9340 8420 28700 28500 490 1630 33200 5890 1020 1330 1210 4080 1920 26600 6670 870 800 4600 1550 7540 9100 3770 880 3960 570 4380 100000 570 830 3910 1400 3800 440 340 2290 1860 520 5080 1920 420 2570 6540 2310 5280 980 700 650 810 11400 330 390 29500 2240 5190 440 320 1580 1190 1090 520
2020-05-28 37800 127000 9350 8450 28800 28500 500 1640 33300 5900 1030 1340 1220 4110 1920 27700 6730 920 820 4760 1590 7570 9600 3860 880 4090 600 4400 100000 580 830 3960 1410 3830 450 350 2310 1870 540 5140 1930 430 2580 6590 2340 5310 1020 710 650 820 11400 330 400 29600 2310 5230 440 320 1590 1200 1090 520
2020-05-29 38000 127000 9360 8490 28900 28600 500 1650 33400 5920 1040 1340 1230 4140 1920 28700 6780 960 840 4930 1640 7600 10100 3950 890 4220 640 4410 101000 590 840 4000 1420 3850 450 350 2310 1880 560 5190 1940 430 2590 6640 2370 5340 1050 710 660 830 11500 340 400 29700 2410 5270 450 320 1610 1220 1100 530
2020-05-30 38200 128000 9380 8520 29000 28600 500 1650 33500 5950 1050 1350 1240 4160 1930 29800 6830 1000 860 5070 1690 7630 10500 4030 890 4360 670 4430 102000 600 850 4040 1430 3880 460 350 2330 1890 580 5250 1960 430 2600 6690 2400 5370 1080 720 670 860 11600 340 400 29800 2490 5320 450 330 1620 1230 1100 530
2020-05-31 38300 128000 9380 8540 29100 28700 510 1660 33600 5960 1050 1350 1240 4190 1930 30800 6870 1040 880 5230 1730 7640 10900 4110 900 4480 710 4450 103000 600 860 4080 1430 3900 460 350 2350 1900 600 5310 1970 430 2610 6730 2430 5400 1120 730 670 870 11600 340 400 29900 2570 5360 450 330 1630 1240 1110 530

Peak increase in estimated trend of Deaths in Europe 2020-05-22

UKEUBEDEESFRHUIEITNLPLPTROSECH
Peak date04-0904-0704-1004-1504-0204-0704-1804-2403-2804-0704-2404-1105-0104-2204-04
Peak daily increment 1012 3237 359 275 901 1016 14 102 829 161 29 33 25 117 62
Days from 100 to peak 22 34 17 23 20 23 5 22 24 18 18 13 29 25 12
Days from peak/2 to peak 16 24 15 22 16 17 25 25 19 18 28 19 37 27 17
Days since peak 43 45 42 37 50 45 34 28 55 45 28 41 21 30 48

Peak increase in estimated trend of Deaths 2020-05-22

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 -- -- -- --03-19 --05-1804-12 -- --04-1904-1505-0605-0804-2304-2404-2004-2505-1405-0704-20 --05-1304-2904-1604-1404-2504-2904-22 --05-0805-06 --04-3005-1304-1204-07 --05-0704-2904-0305-1404-2204-0604-05
Peak daily increment 178 149 135 29 125 2276 18 40 88 54 118 13 11 54 45 126 50 13 75 255 65 147 21 24 309 11 11 1069 175 21 10 39 56 32 13
Days from 100 to peak 34 14 41 11 22 30 23 26 26 21 18 5 20 36 20 43 25 1 18 24 22 25 27 23 29 16 0 18 34 14 -8 40 7 14 -3
Days from peak/2 to peak 35 16 44 25 24 21 37 36 30 28 20 26 45 42 25 44 32 20 21 23 26 26 41 38 29 41 17 13 35 33 5 49 13 29 11
Days since peak 17 64 4 40 33 37 16 14 29 28 32 27 8 15 32 9 23 36 38 27 23 30 14 16 22 9 40 45 15 23 49 8 30 46 47

Initial visual evaluation of forecasts of Deaths