COVID-19 short-term forecasts Confirmed 2020-04-08


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 The documentation that is provided is still in progress and not peer reviewed.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.
[2020-04-08] 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.

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.

Confirmed count average forecast Europe (bold black line in graphs) 2020-04-09 to 2020-04-13

DateUKEUATBEBSCZDEDKESFIFRIEITNLPLPTROSENOCH
2020-04-08 60733 634604 12942 23403 5131 5312 113296 5402 148220 2487 112950 6074 139422 20549 5205 13141 4761 8419 6086 23280
2020-04-09 67000 662000 13200 24700 5290 5550 118000 5800 154000 2670 121000 6520 143000 21500 5640 13800 5120 9100 6210 24000
2020-04-10 74000 691000 13500 25900 5470 5780 122000 6270 159000 2870 132000 7040 147000 22500 6130 14400 5500 9700 6360 24700
2020-04-11 82000 720000 13700 27300 5660 6020 126000 6780 165000 3080 144000 7600 151000 23500 6670 15000 5900 10400 6510 25400
2020-04-12 91000 750000 14000 28700 5860 6270 131000 7330 170000 3310 158000 8210 155000 24600 7260 15700 6350 11200 6670 26100
2020-04-13 102000 780000 14300 30100 6070 6530 135000 7920 176000 3550 172000 8870 159000 25700 7900 16300 6820 12100 6830 26900

Confirmed count average forecast (bold black line in graphs) 2020-04-09 to 2020-04-13

DateAustraliaBrazilCanadaIranMalaysiaPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-08 6010 16170 19141 64586 4119 3870 429052 19043 5655 8781 15690 10204 15078 17030 16790 20220 47437 149401 9097 2048837
2020-04-09 6120 18100 20700 66700 4280 4040 472000 20800 5910 9800 16800 11400 16700 18500 18500 22000 51000 159000 9500 2187000
2020-04-10 6240 20000 22600 68900 4460 4250 525000 22900 6240 10900 18000 12500 18500 20200 20300 24000 55100 170000 10000 2349000
2020-04-11 6360 22100 24800 71300 4650 4450 585000 25300 6580 12100 19300 13800 20500 22000 22200 26200 59400 182000 10400 2520000
2020-04-12 6480 24400 27100 73700 4840 4670 653000 27900 6930 13500 20700 15100 22800 24000 24400 28700 64000 194000 11000 2701000
2020-04-13 6600 27000 29700 76100 5040 4890 729000 30700 7300 15100 22100 16600 25400 26200 26800 31300 68900 208000 11500 2893000

Confirmed count forecast Europe (bold red line in graphs) 2020-04-09 to 2020-04-13

DateUKEUATBEBSCZDEDKESFIFRIEITNLPLPTROSENOCH
2020-04-08 60733 634604 12942 23403 5131 5312 113296 5402 148220 2487 112950 6074 139422 20549 5205 13141 4761 8419 6086 23280
2020-04-09 67000 661000 13200 24700 5270 5520 118000 5770 153000 2590 118000 6520 143000 21400 5590 13600 5070 9000 6210 23800
2020-04-10 75000 688000 13400 26000 5400 5760 122000 6170 158000 2740 123000 7020 147000 22400 6010 14100 5410 9600 6300 24500
2020-04-11 83000 715000 13700 27400 5530 6000 126000 6600 163000 2900 130000 7550 150000 23400 6470 14700 5760 10300 6410 25200
2020-04-12 92000 742000 13900 28900 5660 6260 130000 7060 167000 3070 137000 8130 154000 24400 6960 15200 6140 11000 6510 25900
2020-04-13 102000 769000 14200 30400 5800 6520 134000 7550 172000 3260 144000 8750 158000 25500 7480 15800 6540 11700 6610 26600

Confirmed count forecast (bold red line in graphs) 2020-04-09 to 2020-04-13

DateAustraliaBrazilCanadaIranMalaysiaPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-08 6010 16170 19141 64586 4119 3870 429052 19043 5655 8781 15690 10204 15078 17030 16790 20220 47437 149401 9097 2048837
2020-04-09 6110 17700 20500 66500 4260 4030 471000 20700 5890 9600 16700 11600 16300 18200 18600 21600 50600 158000 9400 2202000
2020-04-10 6210 19600 22100 68300 4420 4180 519000 22600 6120 10600 17700 13100 17900 19300 20700 23200 53900 168000 9800 2356000
2020-04-11 6310 21700 23800 70100 4580 4330 574000 24700 6340 11700 18700 14700 19500 20400 22900 24700 57300 178000 10200 2519000
2020-04-12 6400 24000 25700 71900 4740 4470 637000 27100 6570 13000 19800 16600 21400 21600 25400 26400 60900 189000 10600 2689000
2020-04-13 6480 26600 27800 73700 4910 4610 708000 29700 6800 14400 21000 18700 23400 22700 28200 28100 64500 201000 11000 2868000

Confirmed count scenario forecast (bold green line in graphs) 2020-04-09 to 2020-04-17

DateUKEUATBEDEDKESFRITNLPLPTNOCHUS-WAAustraliaIranMalaysia
2020-04-08 60733 634604 12942 23403 113296 5402 148220 112950 139422 20549 5205 13141 6086 23280 9097 6010 64586 4119
2020-04-09 63100 656000 13100 24600 116000 5610 153000 118000 143000 21400 5440 13700 6280 23700 9400 6180 66600 4230
2020-04-10 66000 678000 13400 25700 120000 5820 157000 122000 146000 22200 5660 14200 6440 24100 9800 6340 68700 4330
2020-04-11 68000 697000 13500 27100 124000 6010 162000 126000 150000 23000 5900 14700 6600 24500 10100 6480 70500 4440
2020-04-12 70300 718000 13800 28400 127000 6230 167000 127000 153000 23800 6120 15100 6770 24800 10400 6640 72600 4530
2020-04-13 72500 737000 14100 29500 131000 6370 172000 130000 156000 24600 6290 15600 6930 25200 10700 6780 74400 4640
2020-04-14 75100 757000 14400 30900 134000 6600 176000 132000 159000 25500 6520 16000 7090 25600 11000 6910 76400 4720
2020-04-15 77400 774000 14700 32100 137000 6740 180000 133000 161000 26300 6690 16500 7220 25900 11300 7070 78400 4790
2020-04-16 80000 793000 14800 33200 140000 6940 185000 135000 164000 27100 6910 17000 7340 26100 11600 7210 80300 4870
2020-04-17 82100 811000 15100 34200 143000 7110 189000 136000 167000 27700 7080 17400 7460 26300 11800 7380 82100 4970

Peak increase in estimated trend of Confirmed in Europe 2020-04-08

UKEUATBEBSCZDEDKESFIFRIEITNLPLPTROSENOCH
Peak date --04-0403-2603-2903-2503-2703-29 --03-27 -- -- --03-2103-28 --03-3104-03 --03-2703-23
Peak daily increment 31977 873 1435 284 282 5953 8090 5705 1099 841 336 297 1205
Days from 100 to peak 41 18 22 11 14 28 25 27 22 18 20 21 18
Days from peak/2 to peak 25 13 14 11 13 16 14 17 16 14 17 19 12
Days since peak 4 13 10 14 12 10 12 18 11 8 5 12 16

Peak increase in estimated trend of Confirmed 2020-04-08

AustraliaBrazilCanadaIranMalaysiaPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date03-28 --04-0503-3104-0203-31 -- --04-03 --04-03 -- --04-04 --04-0404-0304-0404-02 --
Peak daily increment 404 1494 3106 179 345 391 1169 1876 1653 3761 9409 511
Days from 100 to peak 18 25 35 24 17 19 19 19 17 19 27 26
Days from peak/2 to peak 14 17 28 25 14 16 13 12 13 12 15 23
Days since peak 11 3 8 6 8 5 5 4 4 5 4 6

Initial visual evaluation of forecasts of Confirmed