COVID-19 short-term forecasts Deaths 2020-04-11


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 The documentation that is provided is still in progress and not peer reviewed.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

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, but remains preliminary. Also preliminary is the documentation of the medium term forecasts.

Deaths count average forecast Europe (bold black line in graphs) 2020-04-12 to 2020-04-18

DateUKEUATBEDEDKESFRIEITNLPLPTROSECH
2020-04-11 9875 61986 337 3346 2736 260 16606 13832 320 19468 2643 208 470 291 887 1036
2020-04-12 10900 65100 360 3740 2900 270 17200 14700 350 20000 2780 230 500 320 970 1090
2020-04-13 12000 68400 390 4130 3180 290 17800 15700 370 20600 2930 260 530 340 1080 1140
2020-04-14 13300 71900 410 4570 3490 300 18300 16700 400 21100 3090 290 570 360 1210 1200
2020-04-15 14600 75500 440 5050 3830 310 18900 17700 430 21600 3250 330 610 380 1360 1250
2020-04-16 16100 79400 470 5600 4210 330 19600 18800 470 22200 3420 370 650 410 1530 1320
2020-04-17 17800 83500 510 6210 4630 340 20200 20000 500 22800 3600 430 690 440 1720 1380
2020-04-18 19600 87900 540 6890 5100 360 20900 21300 540 23400 3790 490 730 470 1930 1450

Deaths count average forecast (bold black line in graphs) 2020-04-12 to 2020-04-18

DateBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-11 1124 654 4357 247 20463 632 274 494 445 432 682 806 686 1391 2183 8627 498 133033
2020-04-12 1230 740 4480 270 22700 680 300 560 490 460 760 860 780 1540 2430 9500 520 144000
2020-04-13 1360 820 4600 280 25100 740 330 640 530 500 830 930 880 1680 2690 10400 550 157000
2020-04-14 1490 910 4730 300 27800 810 360 730 580 550 910 1000 990 1840 2970 11400 580 172000
2020-04-15 1630 1010 4860 320 30900 890 390 840 630 600 1000 1070 1120 2010 3270 12400 610 188000
2020-04-16 1790 1120 5000 340 34200 970 420 960 680 660 1100 1150 1270 2210 3610 13600 640 205000
2020-04-17 1970 1240 5140 360 37900 1060 460 1100 740 720 1210 1240 1440 2420 4000 14900 670 224000
2020-04-18 2160 1380 5280 380 42100 1160 500 1270 810 790 1320 1330 1650 2650 4430 16400 710 244000

Deaths count forecast Europe (bold red line in graphs) 2020-04-12 to 2020-04-18

DateUKEUATBEDEDKESFRIEITNLPLPTROSECH
2020-04-11 9875 61986 337 3346 2736 260 16606 13832 320 19468 2643 208 470 291 887 1036
2020-04-12 10800 64700 370 3620 2840 270 17200 14500 340 20000 2770 230 500 320 950 1090
2020-04-13 11900 67600 390 3910 2950 280 17700 15300 370 20500 2910 250 540 350 1010 1140
2020-04-14 13100 70500 420 4260 3070 290 18300 16100 400 21100 3050 270 580 390 1070 1190
2020-04-15 14300 73600 450 4620 3200 300 18900 16900 430 21600 3200 290 620 420 1140 1240
2020-04-16 15700 76800 480 5010 3330 320 19500 17800 460 22200 3360 320 660 450 1220 1300
2020-04-17 17200 80200 520 5440 3470 330 20100 18700 490 22700 3530 350 700 490 1300 1360
2020-04-18 18900 83700 560 5910 3620 340 20800 19600 530 23300 3700 380 750 530 1380 1430

Deaths count forecast (bold red line in graphs) 2020-04-12 to 2020-04-18

DateBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-11 1124 654 4357 247 20463 632 274 494 445 432 682 806 686 1391 2183 8627 498 133033
2020-04-12 1230 750 4470 250 22500 680 290 570 480 460 750 860 750 1510 2380 9400 510 145000
2020-04-13 1330 850 4580 260 24800 730 310 640 520 490 820 910 800 1620 2620 10300 530 157000
2020-04-14 1430 950 4700 270 27300 780 330 730 560 510 900 970 860 1760 2870 11200 550 170000
2020-04-15 1540 1060 4820 280 30000 840 350 830 600 530 990 1030 930 1900 3150 12300 560 184000
2020-04-16 1650 1180 4940 290 33100 900 370 950 640 560 1080 1090 990 2060 3450 13400 580 199000
2020-04-17 1780 1320 5070 300 36400 970 390 1090 680 590 1180 1150 1070 2220 3770 14600 600 215000
2020-04-18 1910 1480 5190 320 40000 1040 410 1240 730 610 1290 1220 1140 2400 4120 16000 610 232000

Deaths count scenario forecast (bold green line in graphs) 2020-04-12 to 2020-04-20

DateEUATDKESITNLPTCH
2020-04-11 61986 337 260 16606 19468 2643 470 1036
2020-04-12 64800 360 270 17300 19900 2770 500 1090
2020-04-13 67400 380 280 17800 20200 2880 540 1150
2020-04-14 69800 400 300 18400 20600 2980 580 1190
2020-04-15 72100 420 310 18900 20900 3080 610 1240
2020-04-16 74400 440 320 19400 21300 3180 640 1280
2020-04-17 76600 460 330 19900 21500 3280 680 1310
2020-04-18 78800 480 340 20300 21800 3350 710 1350
2020-04-19 80800 490 350 20600 22000 3430 740 1380
2020-04-20 82700 510 350 21000 22200 3490 770 1420

Peak increase in estimated trend of Deaths in Europe 2020-04-11

UKEUATBEDEDKESFRIEITNLPLPTROSECH
Peak date --04-04 -- -- --04-0404-0204-06 --03-2804-07 -- -- -- --04-04
Peak daily increment 3209 18 898 1058 827 155 62
Days from 100 to peak 31 3 20 22 24 18 12
Days from peak/2 to peak 21 14 16 15 19 18 17
Days since peak 7 7 9 5 14 4 7

Peak increase in estimated trend of Deaths 2020-04-11

BrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date -- --03-19 -- -- -- -- -- --04-06 --04-07 -- -- -- --04-06 --
Peak daily increment 150 38 58 30
Days from 100 to peak 14 7 11 15
Days from peak/2 to peak 16 15 15 30
Days since peak 23 5 4 5

Initial visual evaluation of forecasts of Deaths