COVID-19 short-term forecasts Deaths 2020-04-09


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 The documentation that is provided is still in progress and not peer reviewed.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.
  • 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) 2020-04-10 to 2020-04-14

DateUKEUATBEDEDKESFRIEITNLPTROSECH
2020-04-09 7978 56384 295 2523 2607 237 15447 12210 263 18279 2396 409 248 793 948
2020-04-10 8900 60500 320 2800 2890 260 16100 13400 290 18800 2570 440 270 910 1010
2020-04-11 9900 65500 340 3080 3170 270 16700 14700 320 19300 2760 470 290 1020 1060
2020-04-12 10900 71000 370 3390 3480 290 17400 16100 350 19900 2970 500 310 1150 1120
2020-04-13 12100 76900 390 3730 3820 310 18000 17600 380 20400 3200 540 340 1300 1180
2020-04-14 13300 83300 420 4110 4190 330 18700 19400 420 21000 3440 570 360 1460 1250

Deaths count average forecast (bold black line in graphs) 2020-04-10 to 2020-04-14

DateBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-09 950 503 4110 203 16478 548 227 380 370 412 534 702 503 1076 1700 7067 478 115839
2020-04-10 1070 570 4230 220 18500 600 260 430 410 460 600 760 580 1190 1920 7900 520 127000
2020-04-11 1190 650 4370 230 20700 660 280 490 450 500 670 830 640 1310 2140 8800 550 139000
2020-04-12 1320 740 4500 250 23200 720 310 550 500 560 740 900 710 1450 2390 9800 590 153000
2020-04-13 1460 840 4650 270 25900 790 330 620 550 610 810 980 790 1600 2670 10900 640 168000
2020-04-14 1620 960 4790 280 29000 860 370 700 600 680 900 1070 880 1760 2980 12200 680 185000

Deaths count forecast Europe (bold red line in graphs) 2020-04-10 to 2020-04-14

DateUKEUATBEDEDKESFRIEITNLPTROSECH
2020-04-09 7978 56384 295 2523 2607 237 15447 12210 263 18279 2396 409 248 793 948
2020-04-10 8800 60500 310 2730 2800 250 16000 13100 300 18700 2520 440 270 930 1000
2020-04-11 9800 65200 330 2990 3020 270 16600 14200 340 19300 2680 480 290 1090 1050
2020-04-12 10800 70100 360 3280 3280 290 17200 15300 370 19800 2840 510 320 1280 1100
2020-04-13 12000 75500 380 3590 3550 310 17800 16400 410 20300 3020 550 340 1480 1160
2020-04-14 13200 81400 400 3930 3840 340 18400 17600 460 20800 3210 590 370 1720 1210

Deaths count forecast (bold red line in graphs) 2020-04-10 to 2020-04-14

DateBrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-09 950 503 4110 203 16478 548 227 380 370 412 534 702 503 1076 1700 7067 478 115839
2020-04-10 1040 560 4230 230 18000 580 270 400 400 450 580 730 550 1150 1750 8200 530 127000
2020-04-11 1150 620 4340 250 19900 620 310 430 430 490 640 760 590 1220 1860 9300 570 138000
2020-04-12 1280 690 4460 270 21900 660 350 470 460 530 700 790 640 1300 1970 10500 620 151000
2020-04-13 1410 770 4590 300 24200 710 400 500 500 580 770 820 690 1380 2090 11800 670 164000
2020-04-14 1560 850 4710 320 26700 750 450 540 540 630 840 860 750 1470 2220 13400 720 179000

Deaths count scenario forecast (bold green line in graphs) 2020-04-10 to 2020-04-18

DateEUATDKESITNLPTCH
2020-04-09 56384 295 237 15447 18279 2396 409 948
2020-04-10 58900 320 250 16100 18700 2530 440 1000
2020-04-11 61400 340 260 16700 19200 2650 470 1060
2020-04-12 63900 360 280 17300 19600 2770 490 1110
2020-04-13 66300 380 290 17900 20000 2880 520 1160
2020-04-14 68600 400 300 18400 20400 3020 540 1200
2020-04-15 70800 430 310 18900 20700 3100 560 1220
2020-04-16 73100 450 320 19500 20900 3180 590 1250
2020-04-17 75300 480 330 20000 21200 3270 610 1280
2020-04-18 77300 500 340 20300 21400 3360 630 1300

Peak increase in estimated trend of Deaths in Europe 2020-04-09

UKEUATBEDEDKESFRIEITNLPTROSECH
Peak date -- -- -- -- -- --04-02 -- --03-28 -- -- -- --04-04
Peak daily increment 904 825 61
Days from 100 to peak 20 24 12
Days from peak/2 to peak 16 19 17
Days since peak 7 12 5

Peak increase in estimated trend of Deaths 2020-04-09

BrazilCanadaIranPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date -- --03-19 -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Peak daily increment 150
Days from 100 to peak 14
Days from peak/2 to peak 16
Days since peak 21

Initial visual evaluation of forecasts of Deaths