COVID-19 short-term forecasts Deaths 2020-04-14


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 The documentation that is provided is still in progress and not peer reviewed.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.
    Bird and Nielsen look into nowcasting death counts in England.
  • 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) 2020-04-15 to 2020-04-19

DateUKEUATBEDEDKESFRIEITNLPLPTROSECH
2020-04-14 12107 69270 384 4157 3294 299 18056 15729 406 21067 2945 263 567 351 1033 1174
2020-04-15 13000 71800 400 4520 3450 310 18500 16400 440 21600 3050 280 600 370 1090 1220
2020-04-16 14200 74600 420 4950 3600 320 19000 17100 470 22100 3160 310 640 400 1150 1270
2020-04-17 15400 77500 440 5420 3750 330 19600 17800 500 22600 3280 330 680 430 1200 1320
2020-04-18 16700 80500 460 5930 3910 340 20100 18500 530 23200 3400 350 720 460 1260 1370
2020-04-19 18200 83600 490 6480 4080 360 20700 19200 570 23700 3520 380 760 490 1320 1430

Deaths count average forecast (bold black line in graphs) 2020-04-15 to 2020-04-19

DateBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-14 1532 899 4683 335 25832 778 329 671 570 517 878 1013 957 1766 2805 10834 552 156136
2020-04-15 1690 990 4790 360 28100 830 350 740 620 540 960 1080 1060 1920 3070 11800 570 165000
2020-04-16 1820 1080 4900 390 30700 880 380 800 660 560 1050 1130 1160 2080 3330 12800 590 175000
2020-04-17 1970 1180 5010 420 33600 940 410 870 700 580 1160 1190 1270 2240 3610 13900 600 185000
2020-04-18 2140 1290 5130 460 36800 1000 440 940 740 600 1270 1250 1400 2420 3910 15100 620 196000
2020-04-19 2310 1400 5240 490 40200 1060 470 1020 790 620 1390 1310 1530 2620 4250 16400 640 207000

Deaths count forecast Europe (bold red line in graphs) 2020-04-15 to 2020-04-19

DateUKEUATBEDEDKESFRIEITNLPLPTROSECH
2020-04-14 12107 69270 384 4157 3294 299 18056 15729 406 21067 2945 263 567 351 1033 1174
2020-04-15 13000 71600 400 4570 3410 310 18400 16300 430 21500 3030 280 600 380 1100 1210
2020-04-16 14100 74000 420 4950 3530 320 18800 16900 470 22000 3120 300 630 400 1160 1250
2020-04-17 15200 76500 430 5350 3640 330 19200 17500 500 22500 3220 320 670 430 1210 1290
2020-04-18 16400 79000 450 5780 3760 340 19500 18100 530 23100 3310 340 710 460 1260 1320
2020-04-19 17700 81700 470 6250 3870 350 19900 18800 570 23600 3410 360 750 500 1320 1360

Deaths count forecast (bold red line in graphs) 2020-04-15 to 2020-04-19

DateBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-14 1532 899 4683 335 25832 778 329 671 570 517 878 1013 957 1766 2805 10834 552 156136
2020-04-15 1740 1000 4780 370 28400 820 350 750 640 530 950 1060 1050 1930 3170 11900 570 165000
2020-04-16 1980 1110 4890 400 31200 870 370 830 710 560 1030 1130 1160 2100 3570 13000 590 174000
2020-04-17 2260 1230 4990 430 34300 910 390 920 790 580 1110 1210 1270 2290 4010 14200 610 183000
2020-04-18 2560 1360 5090 460 37700 960 410 1010 880 610 1190 1290 1390 2480 4510 15600 630 193000
2020-04-19 2900 1510 5190 500 41500 1010 430 1110 990 630 1280 1370 1520 2700 5070 17100 650 203000

Deaths count scenario forecast (bold green line in graphs) 2020-04-15 to 2020-04-23

DateEUATDKESITNLPTCH
2020-04-14 69270 384 299 18056 21067 2945 567 1174
2020-04-15 71500 400 310 18600 21500 3050 600 1220
2020-04-16 73800 410 320 19100 21900 3150 630 1260
2020-04-17 76000 430 330 19500 22400 3250 650 1300
2020-04-18 78100 440 340 19900 22800 3340 680 1340
2020-04-19 80000 450 340 20300 23200 3430 710 1370
2020-04-20 81800 460 350 20600 23700 3520 740 1410
2020-04-21 83700 480 360 20900 24100 3600 760 1440
2020-04-22 85300 490 360 21100 24500 3680 780 1470
2020-04-23 86800 500 370 21400 24800 3740 810 1490

Peak increase in estimated trend of Deaths in Europe 2020-04-14

UKEUATBEDEDKESFRIEITNLPLPTROSECH
Peak date --04-0704-10 --04-0804-0403-3004-07 --03-2804-0704-08 -- --04-0904-04
Peak daily increment 3258 23 231 18 896 1041 826 161 20 81 61
Days from 100 to peak 34 11 16 3 17 23 24 18 2 12 12
Days from peak/2 to peak 24 19 16 14 13 16 19 18 15 15 17
Days since peak 7 4 6 10 15 7 17 7 6 5 10

Peak increase in estimated trend of Deaths 2020-04-14

BrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date -- --03-19 -- --04-08 -- -- --04-07 -- -- -- -- -- --04-06 --
Peak daily increment 150 49 38 29
Days from 100 to peak 14 11 8 15
Days from peak/2 to peak 16 18 16 30
Days since peak 26 6 7 8

Initial visual evaluation of forecasts of Deaths