COVID-19 short-term forecasts Deaths 2020-05-01


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 The documentation that is provided is still in progress and not peer reviewed.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

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) 2020-05-02 to 2020-05-08

DateUKEUATBECZDEDKESFIFRHUIEITNLPLPTROSENOCH
2020-05-01 27510 105446 589 7703 240 6736 460 24543 218 24563 323 1265 28236 4893 651 1007 744 2653 210 1754
2020-05-02 27900 107000 600 7810 240 6860 460 24800 220 24900 330 1300 28500 4930 670 1020 770 2690 210 1780
2020-05-03 28500 108000 600 7930 240 6990 470 25000 220 25200 330 1340 28800 4990 700 1050 800 2740 210 1800
2020-05-04 29200 110000 610 8060 250 7130 480 25300 230 25500 340 1380 29100 5060 720 1070 830 2800 210 1820
2020-05-05 29800 111000 620 8180 250 7270 480 25600 230 25800 350 1420 29400 5140 750 1090 870 2870 220 1850
2020-05-06 30500 113000 630 8300 250 7410 490 25900 240 26100 360 1460 29700 5210 780 1110 900 2940 220 1870
2020-05-07 31200 114000 640 8430 250 7550 500 26200 240 26500 370 1500 30000 5290 810 1130 940 3010 220 1900
2020-05-08 31900 116000 650 8570 260 7700 510 26400 250 26800 380 1550 30400 5370 840 1150 970 3090 220 1920

Deaths count average forecast (bold black line in graphs) 2020-05-02 to 2020-05-08

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-05-01 6412 3537 1223 800 6091 1972 579 3258 64943 2147 818 2339 231 1313 1153 2464 1927 3716 3866 7538 18610 829 311844
2020-05-02 6900 3610 1300 810 6160 2090 590 3350 66200 2250 830 2390 240 1360 1180 2570 1950 3830 3960 7900 18900 840 320000
2020-05-03 7400 3770 1380 820 6230 2240 610 3440 68000 2360 850 2460 250 1420 1210 2710 1980 4000 4060 8400 19200 860 328000
2020-05-04 8000 3930 1460 840 6300 2400 620 3530 69900 2480 880 2530 260 1490 1250 2860 2030 4170 4170 8800 19500 870 336000
2020-05-05 8600 4100 1550 850 6370 2570 640 3630 71800 2600 900 2610 270 1550 1290 3020 2070 4350 4280 9300 19900 880 344000
2020-05-06 9300 4280 1650 860 6440 2750 650 3730 73800 2730 930 2680 280 1630 1330 3190 2120 4540 4390 9800 20200 900 353000
2020-05-07 10000 4470 1750 870 6510 2950 670 3830 75800 2870 950 2760 300 1700 1370 3360 2160 4740 4510 10300 20500 910 361000
2020-05-08 10700 4670 1860 890 6590 3150 690 3940 78000 3010 980 2850 310 1780 1420 3550 2210 4950 4620 10900 20900 930 369000

Deaths count forecast Europe (bold red line in graphs) 2020-05-02 to 2020-05-08

DateUKEUATBECZDEDKESFIFRHUIEITNLPLPTROSENOCH
2020-05-01 27510 105446 589 7703 240 6736 460 24543 218 24563 323 1265 28236 4893 651 1007 744 2653 210 1754
2020-05-02 28100 107000 600 7800 240 6850 470 24800 220 24900 330 1290 28500 4970 660 1030 770 2720 210 1770
2020-05-03 28700 108000 600 7910 250 6980 470 25000 230 25100 340 1330 28800 5040 680 1040 800 2790 210 1790
2020-05-04 29300 109000 610 8000 250 7100 480 25200 230 25400 350 1350 29100 5120 690 1060 830 2860 210 1810
2020-05-05 29900 111000 620 8100 250 7220 490 25400 240 25700 360 1380 29300 5200 710 1080 860 2930 210 1830
2020-05-06 30600 112000 630 8200 260 7350 490 25500 240 26000 360 1410 29600 5270 720 1100 900 3010 210 1850
2020-05-07 31200 113000 630 8290 260 7480 500 25700 240 26300 370 1440 29900 5350 730 1120 930 3080 210 1870
2020-05-08 31900 114000 640 8390 260 7610 510 25900 250 26600 380 1470 30200 5430 750 1140 970 3160 220 1890

Deaths count forecast (bold red line in graphs) 2020-05-02 to 2020-05-08

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-05-01 6412 3537 1223 800 6091 1972 579 3258 64943 2147 818 2339 231 1313 1153 2464 1927 3716 3866 7538 18610 829 311844
2020-05-02 7000 3690 1300 810 6160 2130 590 3340 66800 2280 830 2410 240 1370 1190 2610 1980 3870 3960 8200 18900 840 320000
2020-05-03 7500 3850 1380 820 6220 2300 610 3430 68600 2430 860 2480 250 1430 1240 2760 2020 4020 4040 8800 19200 850 328000
2020-05-04 8100 4020 1470 830 6290 2470 620 3510 70500 2590 880 2550 260 1490 1280 2930 2070 4180 4130 9400 19500 870 335000
2020-05-05 8800 4190 1560 840 6350 2650 640 3600 72400 2760 910 2630 270 1550 1320 3110 2110 4340 4220 10100 19800 880 343000
2020-05-06 9400 4370 1650 840 6420 2840 650 3690 74300 2940 940 2700 290 1620 1370 3300 2160 4510 4310 10800 20000 890 351000
2020-05-07 10200 4560 1760 850 6490 3050 670 3780 76300 3130 960 2780 300 1680 1420 3500 2200 4690 4410 11600 20300 900 358000
2020-05-08 11000 4760 1860 860 6550 3280 680 3870 78400 3330 990 2850 310 1750 1470 3710 2250 4870 4500 12400 20600 920 365000

Deaths count scenario forecast (bold purple line in graphs) 2020-05-02 to 2020-05-10

DateUKEUATBECZDEDKESFIFRHUIEITNLPLPTROSENOCHBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-05-01 27510 105446 589 7703 240 6736 460 24543 218 24563 323 1265 28236 4893 651 1007 744 2653 210 1754 6412 3537 1223 800 6091 1972 579 3258 64943 2147 818 2339 231 1313 1153 2464 1927 3716 3866 7538 18610 829 311844
2020-05-02 28000 107000 600 7790 240 6840 470 24900 220 24900 330 1280 28500 4940 670 1020 760 2700 210 1770 7000 3580 1290 810 6140 2120 590 3340 66300 2230 820 2390 240 1360 1190 2570 1930 3860 3950 7800 18800 840 320000
2020-05-03 28500 107000 600 7880 240 6940 470 25100 230 25100 340 1290 28800 5010 690 1040 790 2770 210 1790 7600 3720 1360 820 6190 2250 600 3410 68100 2320 840 2450 250 1390 1230 2680 1980 4000 4030 8080 19100 850 328000
2020-05-04 29100 108000 610 7960 250 7060 480 25200 230 25300 350 1310 29000 5080 710 1050 810 2840 220 1810 8200 3850 1430 830 6230 2380 610 3490 69700 2390 860 2510 260 1430 1270 2760 2010 4140 4120 8330 19300 870 334000
2020-05-05 29600 109000 620 8040 250 7180 480 25400 240 25500 350 1330 29200 5140 720 1060 830 2900 220 1820 8900 3980 1500 830 6270 2500 620 3560 71200 2470 880 2570 270 1470 1310 2850 2040 4270 4190 8570 19500 880 341000
2020-05-06 30100 110000 620 8120 250 7280 490 25600 240 25600 360 1350 29400 5200 730 1080 850 2960 220 1830 9600 4100 1560 840 6310 2620 620 3620 72600 2530 890 2630 280 1500 1340 2930 2080 4410 4280 8770 19700 890 347000
2020-05-07 30400 111000 630 8190 250 7380 490 25800 250 25800 370 1370 29700 5250 750 1090 860 3010 220 1850 10100 4220 1620 850 6350 2740 630 3680 73900 2590 910 2680 280 1530 1360 3020 2110 4540 4350 8950 19800 900 352000
2020-05-08 30800 112000 640 8270 250 7500 500 25900 250 25900 380 1390 29900 5300 770 1100 880 3050 230 1860 10800 4320 1680 860 6390 2860 640 3730 75000 2650 930 2740 290 1560 1390 3100 2130 4670 4420 9170 20000 900 356000
2020-05-09 31100 113000 640 8340 260 7610 500 26100 260 26000 380 1410 30100 5340 780 1120 900 3080 230 1880 11500 4430 1740 870 6420 2980 640 3780 76000 2700 940 2780 300 1580 1410 3160 2150 4790 4480 9380 20100 910 361000
2020-05-10 31500 114000 640 8400 260 7690 500 26200 260 26100 390 1430 30300 5390 800 1130 920 3130 230 1890 12100 4540 1790 870 6450 3070 640 3810 77000 2740 960 2820 310 1610 1430 3230 2170 4890 4550 9560 20300 920 365000

Peak increase in estimated trend of Deaths in Europe 2020-05-01

UKEUATBECZDEDKESFIFRHUIEITNLPLPTROSENOCH
Peak date04-0904-0704-0804-1004-0904-1504-0404-0204-2104-0704-1804-2403-2804-0704-2504-11 --04-2204-0804-04
Peak daily increment 1014 3241 23 364 11 280 18 902 18 1014 15 102 826 163 29 33 115 9 61
Days from 100 to peak 22 34 9 17 0 23 3 20 1 23 5 22 24 18 19 13 25 -1 12
Days from peak/2 to peak 16 24 17 15 14 22 14 16 23 17 25 25 19 18 29 19 27 21 17
Days since peak 22 24 23 21 22 16 27 29 10 24 13 7 34 24 6 20 9 23 27

Peak increase in estimated trend of Deaths 2020-05-01

BrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date -- -- --04-1403-19 --04-1204-1904-15 --04-2404-2004-25 -- -- --04-1504-2404-21 --04-0904-0604-17
Peak daily increment 39 150 28 125 2279 56 116 13 75 179 144 777 30 9904
Days from 100 to peak 16 14 11 22 29 21 18 5 19 23 24 18 15 82
Days from peak/2 to peak 26 16 25 24 21 28 20 26 22 23 25 14 30 54
Days since peak 17 43 19 12 16 7 11 6 16 7 10 22 25 14

Initial visual evaluation of forecasts of Deaths