COVID-19 short-term forecasts Confirmed 2020-04-10


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 The documentation that is provided is still in progress and not peer reviewed.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

Confirmed count average forecast Europe (bold black line in graphs) 2020-04-11 to 2020-04-15

DateUKEUATBEBSCZDEDKESFIFRGRIEITNLPLPTROSENOCH
2020-04-10 73758 689440 13555 26667 5480 5732 122171 5819 158273 2769 124869 2011 8089 147577 23097 5955 15472 5467 9685 6314 24551
2020-04-11 81000 715000 13800 28600 5640 5950 126000 6090 163000 2950 132000 2070 9000 151000 24500 6370 16700 5890 10400 6410 25200
2020-04-12 89000 741000 14100 30700 5790 6190 130000 6440 168000 3170 140000 2120 9700 155000 26100 6840 17800 6350 11200 6540 25800
2020-04-13 98000 767000 14400 33000 5940 6430 135000 6810 172000 3400 150000 2170 10500 159000 27800 7350 19100 6840 12000 6670 26500
2020-04-14 107000 793000 14600 35500 6090 6680 139000 7210 177000 3650 160000 2220 11400 163000 29700 7890 20400 7380 12900 6800 27200
2020-04-15 118000 819000 14900 38200 6250 6940 143000 7630 182000 3920 171000 2270 12400 166000 31600 8480 21900 7960 13900 6940 27800

Confirmed count average forecast (bold black line in graphs) 2020-04-11 to 2020-04-15

DateAustraliaBrazilCanadaIranMalaysiaPhilippinesSouth AfricaUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-10 6215 19638 22059 68192 4346 4195 2003 496535 21366 6510 10538 17960 11859 17887 19253 20974 22646 54588 170512 9887 2300190
2020-04-11 6310 21800 23600 69900 4480 4340 2080 534000 22700 6850 11600 19000 13000 19800 20400 23100 24000 58000 181000 10300 2434000
2020-04-12 6400 24200 25300 71700 4640 4480 2150 574000 24100 7130 12600 20100 14300 22000 21700 25300 25500 61500 192000 10700 2572000
2020-04-13 6490 26800 27200 73500 4800 4630 2220 619000 25600 7420 13800 21200 15800 24500 23200 27700 27100 65100 203000 11100 2715000
2020-04-14 6570 29800 29200 75200 4970 4790 2300 666000 27200 7720 15000 22400 17400 27200 24700 30400 28700 68900 214000 11500 2864000
2020-04-15 6660 33100 31300 77100 5140 4940 2370 718000 28900 8030 16400 23600 19200 30300 26300 33300 30500 72900 227000 12000 3018000

Confirmed count forecast Europe (bold red line in graphs) 2020-04-11 to 2020-04-15

DateUKEUATBEBSCZDEDKESFIFRGRIEITNLPLPTROSENOCH
2020-04-10 73758 689440 13555 26667 5480 5732 122171 5819 158273 2769 124869 2011 8089 147577 23097 5955 15472 5467 9685 6314 24551
2020-04-11 82000 710000 13800 28600 5630 5920 126000 6050 163000 2870 130000 2060 8600 151000 24600 6320 17200 5960 10300 6410 25100
2020-04-12 91000 734000 14100 30600 5790 6110 130000 6290 167000 3020 137000 2110 9400 155000 26300 6720 19100 6450 10900 6500 25800
2020-04-13 101000 757000 14300 32800 5940 6310 134000 6540 171000 3170 143000 2160 10400 158000 28000 7150 21100 6970 11600 6590 26400
2020-04-14 113000 779000 14600 35300 6100 6520 138000 6790 175000 3330 149000 2210 11400 162000 29900 7600 23400 7540 12400 6680 27000
2020-04-15 125000 802000 14900 37900 6260 6730 143000 7050 180000 3500 156000 2260 12600 165000 31900 8070 26000 8150 13200 6770 27700

Confirmed count forecast (bold red line in graphs) 2020-04-11 to 2020-04-15

DateAustraliaBrazilCanadaIranMalaysiaPhilippinesSouth AfricaUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-10 6215 19638 22059 68192 4346 4195 2003 496535 21366 6510 10538 17960 11859 17887 19253 20974 22646 54588 170512 9887 2300190
2020-04-11 6300 21900 23400 69800 4470 4330 2080 530000 22500 6800 11800 18700 13300 20100 20300 22500 23800 57400 180000 10200 2442000
2020-04-12 6380 24500 24900 71500 4600 4450 2150 563000 23700 7090 13100 19700 14800 22200 21400 24300 24900 60600 190000 10500 2586000
2020-04-13 6450 27300 26600 73100 4730 4580 2220 598000 24800 7390 14500 20600 16400 24600 22500 26200 26100 63700 200000 10800 2735000
2020-04-14 6530 30400 28500 74700 4870 4700 2290 635000 26000 7700 16000 21600 18200 27200 23500 28300 27300 67000 210000 11200 2889000
2020-04-15 6600 33900 30500 76400 5010 4820 2370 674000 27200 8020 17700 22600 20100 30100 24600 30600 28600 70400 221000 11500 3050000

Confirmed count scenario forecast (bold green line in graphs) 2020-04-11 to 2020-04-19

DateUKEUATBEDEDKESFRITNLPLPTNOCHUS-WAAustraliaIranMalaysia
2020-04-10 73758 689440 13555 26667 122171 5819 158273 124869 147577 23097 5955 15472 6314 24551 9887 6215 68192 4346
2020-04-11 77000 704000 13800 27800 126000 6020 163000 128000 151000 23800 6180 15500 6450 25200 10200 6320 70000 4450
2020-04-12 81000 724000 14100 29100 129000 6300 168000 131000 154000 24600 6470 16200 6580 25800 10500 6450 71600 4530
2020-04-13 85000 743000 14400 30200 133000 6400 173000 136000 157000 25400 6620 16900 6710 26400 10800 6550 73400 4610
2020-04-14 88000 763000 14500 31200 136000 6660 177000 140000 161000 26200 6910 17500 6820 27000 11100 6690 75100 4690
2020-04-15 92000 782000 14800 32400 139000 6780 181000 144000 164000 26800 7080 18000 6940 27500 11400 6830 76800 4760
2020-04-16 95000 801000 14900 33600 142000 6990 186000 148000 167000 27600 7330 18500 7050 28000 11700 6930 78500 4840
2020-04-17 98000 819000 15100 34300 146000 7180 190000 152000 169000 28300 7450 19000 7150 28500 12000 7020 80200 4910
2020-04-18 101000 836000 15200 34800 149000 7280 194000 156000 171000 28900 7530 19400 7240 29000 12200 7100 81900 5000
2020-04-19 104000 854000 15400 35500 152000 7450 199000 159000 174000 29400 7800 19900 7290 29400 12500 7170 83600 5080

Peak increase in estimated trend of Confirmed in Europe 2020-04-10

UKEUATBEBSCZDEDKESFIFRGRIEITNLPLPTROSENOCH
Peak date --04-0403-26 --03-2503-2803-2804-0703-27 --04-0704-01 --03-21 --04-05 -- -- --03-2703-23
Peak daily increment 32636 882 287 285 6008 344 8144 10067 95 5704 403 299 1205
Days from 100 to peak 41 18 11 15 27 28 25 37 21 27 21 21 18
Days from peak/2 to peak 25 13 11 14 15 28 14 22 24 17 19 19 12
Days since peak 6 15 16 13 13 3 14 3 9 20 5 14 18

Peak increase in estimated trend of Confirmed 2020-04-10

AustraliaBrazilCanadaIranMalaysiaPhilippinesSouth AfricaUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date03-28 --04-0503-2904-0203-3103-2604-0604-0404-03 --04-03 -- --04-04 --04-0404-0304-0404-02 --
Peak daily increment 409 1488 3092 180 351 195 31000 1349 394 1187 1771 1663 3850 9686 516
Days from 100 to peak 18 25 33 24 17 8 34 27 19 19 19 17 19 27 26
Days from peak/2 to peak 14 17 26 25 13 8 17 18 16 13 13 13 12 15 23
Days since peak 13 5 12 8 10 15 4 6 7 7 6 6 7 6 8

Initial visual evaluation of forecasts of Confirmed