COVID-19 short-term forecasts Deaths 2020-04-20


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 The documentation that is provided is still in progress and not peer reviewed.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

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) 2020-04-21 to 2020-04-27

DateUKEUATBEDEDKESFRIEITNLPLPTROSECH
2020-04-20 16509 85325 470 5828 4862 364 20852 20265 687 24114 3751 380 735 478 1580 1429
2020-04-21 17100 87000 480 6060 5130 370 21200 20800 740 24600 3850 400 760 500 1650 1470
2020-04-22 17900 90000 500 6330 5460 390 21600 21500 790 25100 3990 420 790 520 1740 1510
2020-04-23 18600 92000 510 6600 5800 400 22000 22200 850 25600 4130 440 820 540 1840 1550
2020-04-24 19400 95000 530 6880 6170 410 22400 23000 900 26100 4270 460 850 550 1950 1590
2020-04-25 20300 98000 540 7170 6570 420 22800 23800 970 26600 4420 480 880 570 2070 1630
2020-04-26 21100 100000 560 7490 6990 430 23300 24600 1030 27100 4570 500 920 590 2190 1680
2020-04-27 22100 103000 570 7810 7430 450 23700 25400 1100 27600 4740 530 950 620 2330 1720

Deaths count average forecast (bold black line in graphs) 2020-04-21 to 2020-04-27

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-20 2587 1725 592 590 5209 686 428 2140 42094 1223 448 1331 822 767 1359 1328 1809 2466 4377 14347 661 218522
2020-04-21 2750 1890 630 610 5290 750 440 2280 44500 1280 470 1440 860 810 1430 1370 1970 2560 4600 14800 680 228000
2020-04-22 2970 2060 660 640 5370 820 460 2420 47900 1360 480 1550 910 860 1520 1430 2160 2680 4910 15300 700 241000
2020-04-23 3200 2240 690 660 5450 890 470 2580 51500 1450 500 1670 960 900 1620 1480 2360 2810 5240 15900 710 255000
2020-04-24 3450 2450 730 690 5530 970 490 2750 55400 1540 520 1790 1010 950 1720 1540 2590 2950 5590 16400 730 269000
2020-04-25 3730 2670 770 710 5620 1050 500 2930 59700 1640 530 1930 1070 1000 1840 1600 2850 3090 5980 16900 750 284000
2020-04-26 4040 2920 810 740 5700 1140 520 3110 64400 1750 550 2090 1130 1050 1960 1660 3130 3240 6400 17500 770 299000
2020-04-27 4370 3200 850 770 5790 1250 530 3320 69400 1860 570 2250 1200 1110 2090 1730 3440 3400 6850 18100 790 315000

Deaths count forecast Europe (bold red line in graphs) 2020-04-21 to 2020-04-27

DateUKEUATBEDEDKESFRIEITNLPLPTROSECH
2020-04-20 16509 85325 470 5828 4862 364 20852 20265 687 24114 3751 380 735 478 1580 1429
2020-04-21 17100 87200 480 6050 5180 370 21200 20800 730 24500 3850 400 760 490 1650 1460
2020-04-22 17800 89200 500 6260 5510 380 21600 21400 790 25000 3950 420 780 510 1720 1500
2020-04-23 18400 91200 510 6480 5840 390 21900 21900 850 25400 4050 430 810 530 1790 1540
2020-04-24 19100 93200 520 6690 6200 400 22300 22500 910 25900 4150 450 830 560 1860 1570
2020-04-25 19800 95300 540 6910 6580 410 22700 23100 980 26300 4250 470 860 580 1940 1610
2020-04-26 20600 97500 550 7130 6990 420 23100 23700 1050 26800 4360 490 880 600 2020 1650
2020-04-27 21300 99700 570 7350 7420 440 23500 24300 1130 27300 4470 520 910 630 2100 1690

Deaths count forecast (bold red line in graphs) 2020-04-21 to 2020-04-27

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-20 2587 1725 592 590 5209 686 428 2140 42094 1223 448 1331 822 767 1359 1328 1809 2466 4377 14347 661 218522
2020-04-21 2730 1890 620 610 5290 730 440 2280 44200 1280 460 1420 860 800 1420 1370 1950 2550 4590 14800 680 229000
2020-04-22 2890 2060 660 630 5360 770 460 2420 46500 1340 480 1550 890 850 1490 1410 2110 2630 4820 15200 700 240000
2020-04-23 3050 2250 690 650 5440 820 470 2580 48900 1400 500 1690 930 890 1550 1440 2270 2710 5040 15600 710 251000
2020-04-24 3220 2460 730 670 5510 870 480 2740 51400 1470 520 1830 980 950 1620 1480 2450 2800 5280 16100 730 262000
2020-04-25 3400 2680 770 690 5590 920 500 2910 54000 1540 550 1990 1020 1000 1700 1520 2640 2880 5530 16500 750 273000
2020-04-26 3590 2930 810 700 5670 970 510 3100 56700 1610 570 2170 1070 1060 1780 1550 2850 2960 5790 17000 770 285000
2020-04-27 3790 3200 860 720 5740 1030 530 3290 59700 1690 590 2360 1110 1120 1860 1590 3080 3040 6060 17400 790 298000

Deaths count scenario forecast (bold green line in graphs) 2020-04-21 to 2020-04-29

DateUKEUATBEDEDKESFRIEITNLPLPTROSECHBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-20 16509 85325 470 5828 4862 364 20852 20265 687 24114 3751 380 735 478 1580 1429 2587 1725 592 590 5209 686 428 2140 42094 1223 448 1331 822 767 1359 1328 1809 2466 4377 14347 661 218522
2020-04-21 17300 87200 480 6140 5080 370 21200 20900 710 24500 3820 390 760 490 1640 1460 2730 1820 620 620 5270 740 440 2260 44400 1280 460 1330 860 770 1440 1370 1960 2610 4520 14900 670 229000
2020-04-22 17900 88900 490 6380 5290 380 21500 21500 760 24900 3870 410 780 500 1670 1490 2850 1920 650 650 5350 800 460 2370 46100 1330 470 1360 900 780 1510 1400 2090 2700 4730 15500 680 239000
2020-04-23 18500 90400 500 6600 5510 390 21800 22100 800 25300 3930 420 810 520 1700 1520 2980 2030 670 670 5410 860 470 2480 48000 1370 490 1430 950 790 1580 1440 2210 2800 4860 16000 690 249000
2020-04-24 19100 91800 500 6820 5700 400 22000 22600 840 25700 3990 440 830 540 1740 1540 3090 2120 690 690 5470 910 490 2590 48900 1420 510 1500 980 810 1650 1470 2310 2890 5030 16400 700 259000
2020-04-25 19700 93000 510 7040 5900 410 22200 23100 880 26100 4050 450 850 550 1790 1560 3220 2210 720 710 5520 970 500 2690 49700 1460 520 1570 1020 840 1720 1510 2420 2990 5170 16900 710 268000
2020-04-26 20300 94200 520 7270 6080 420 22400 23600 920 26400 4110 460 870 560 1830 1580 3340 2310 740 730 5580 1010 510 2800 50500 1510 530 1640 1050 860 1790 1540 2520 3080 5330 17300 730 276000
2020-04-27 20800 95500 530 7500 6260 420 22600 24100 960 26800 4170 470 890 580 1880 1600 3450 2400 770 750 5620 1050 530 2890 52000 1550 550 1700 1080 880 1850 1570 2610 3160 5480 17700 730 285000
2020-04-28 21400 96200 540 7700 6430 430 22900 24500 1000 27100 4230 480 910 590 1930 1610 3570 2490 790 770 5660 1100 540 2990 53800 1600 560 1750 1120 900 1910 1600 2700 3240 5640 18100 740 293000
2020-04-29 21800 96900 540 7880 6590 440 23100 24900 1030 27400 4290 490 920 600 1960 1630 3670 2580 810 790 5700 1150 550 3080 55600 1650 570 1800 1150 920 1970 1620 2780 3320 5800 18400 750 300000

Peak increase in estimated trend of Deaths in Europe 2020-04-20

UKEUATBEDEDKESFRIEITNLPLPTROSECH
Peak date04-1004-0704-0804-10 --04-0404-0204-07 --03-2804-0704-1504-1104-09 --04-04
Peak daily increment 876 3221 23 357 18 902 1036 826 163 21 33 23 61
Days from 100 to peak 22 34 9 17 3 20 23 24 18 9 13 7 12
Days from peak/2 to peak 16 24 17 15 14 16 16 19 18 21 19 15 17
Days since peak 10 13 12 10 16 18 13 23 13 5 9 11 16

Peak increase in estimated trend of Deaths 2020-04-20

BrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date -- --04-1104-1203-19 --04-12 --04-16 --04-09 --04-14 -- --04-14 --04-15 --04-1104-06 --
Peak daily increment 38 37 150 27 3073 24 48 76 142 784 30
Days from 100 to peak 6 14 14 11 30 6 13 18 18 20 15
Days from peak/2 to peak 15 24 16 25 20 16 20 21 19 16 30
Days since peak 9 8 32 8 4 11 6 6 5 9 14

Initial visual evaluation of forecasts of Deaths