COVID-19 short-term forecasts Confirmed 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 The documentation that is provided is still in progress and not peer reviewed.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.

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

DateUKEUATBEBSCZDEDKESFIFRGRIEITNLPLPTROSENOCH
2020-04-14 93873 755290 14226 31119 5750 6111 131359 6511 172541 3161 130253 2170 11479 162488 27419 7202 17448 6879 11445 6623 25936
2020-04-15 99000 769000 14400 32100 5810 6200 134000 6680 176000 3250 131000 2210 12600 166000 28500 7500 18000 7220 11900 6690 26300
2020-04-16 106000 786000 14600 33400 5900 6330 136000 6840 180000 3360 135000 2250 13900 170000 29600 7850 18700 7620 12400 6770 26700
2020-04-17 113000 802000 14800 34800 5980 6450 139000 7010 184000 3470 138000 2300 15300 173000 30900 8210 19400 8030 13000 6860 27100
2020-04-18 120000 817000 15000 36200 6070 6580 143000 7170 188000 3580 142000 2340 16900 177000 32200 8580 20200 8470 13500 6950 27600
2020-04-19 127000 832000 15200 37700 6150 6710 146000 7340 192000 3690 145000 2390 18600 181000 33500 8970 21000 8940 14100 7040 28100

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

DateAustraliaBrazilCanadaIranMalaysiaPhilippinesSouth AfricaUSUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-14 6415 25262 27034 74877 4987 5223 2415 607670 25758 7941 13989 2058 21620 13913 23247 21518 28163 26844 68824 202208 10694 2701337
2020-04-15 6460 26900 28300 76400 5130 5460 2520 639000 26900 8290 14900 2160 22500 14500 24700 22000 30300 27900 72100 210000 10900 2808000
2020-04-16 6510 28600 29700 78000 5270 5640 2600 676000 28000 8660 15900 2300 23500 15100 26200 22700 33200 28800 75300 220000 11100 2928000
2020-04-17 6560 30300 31200 79600 5410 5830 2680 716000 29200 9040 17000 2430 24400 15700 27800 23300 36300 29800 78600 229000 11300 3051000
2020-04-18 6610 32200 32700 81200 5550 6020 2770 758000 30500 9440 18100 2580 25400 16300 29500 23900 39700 30900 82100 239000 11500 3175000
2020-04-19 6650 34200 34300 82800 5700 6220 2850 804000 31800 9860 19300 2740 26500 17000 31200 24600 43400 31900 85700 250000 11700 3302000

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

DateUKEUATBEBSCZDEDKESFIFRGRIEITNLPLPTROSENOCH
2020-04-14 93873 755290 14226 31119 5750 6111 131359 6511 172541 3161 130253 2170 11479 162488 27419 7202 17448 6879 11445 6623 25936
2020-04-15 99000 769000 14400 31900 5820 6210 133000 6680 175000 3250 133000 2210 13000 166000 28400 7470 18000 7230 11900 6690 26200
2020-04-16 105000 779000 14600 32800 5890 6290 135000 6840 178000 3330 134000 2240 14500 169000 29500 7740 18500 7560 12300 6750 26600
2020-04-17 110000 789000 14800 33600 5950 6370 137000 7000 181000 3420 135000 2280 16000 173000 30600 8010 19100 7920 12700 6810 26900
2020-04-18 116000 797000 14900 34400 6010 6440 139000 7160 184000 3500 137000 2310 17700 176000 31700 8290 19600 8280 13200 6870 27100
2020-04-19 123000 804000 15100 35200 6070 6510 141000 7310 186000 3580 138000 2350 19600 179000 32800 8570 20200 8660 13600 6930 27400

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

DateAustraliaBrazilCanadaIranMalaysiaPhilippinesSouth AfricaUSUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-14 6415 25262 27034 74877 4987 5223 2415 607670 25758 7941 13989 2058 21620 13913 23247 21518 28163 26844 68824 202208 10694 2701337
2020-04-15 6470 26900 28100 76400 5110 5400 2490 635000 26600 8300 14700 2140 22500 14300 24500 22100 30300 27800 71500 210000 10900 2819000
2020-04-16 6510 28400 29400 77900 5250 5620 2590 663000 27700 8650 15500 2230 23300 14800 25900 22700 32800 28900 74800 218000 11000 2932000
2020-04-17 6550 29900 30800 79300 5400 5850 2700 691000 28800 9010 16300 2320 24200 15400 27300 23200 35300 29900 78300 227000 11200 3044000
2020-04-18 6580 31400 32300 80800 5550 6090 2810 720000 29900 9380 17200 2410 25100 15900 28800 23800 38100 31000 81800 235000 11300 3157000
2020-04-19 6620 33000 34000 82300 5700 6340 2920 751000 31100 9770 18000 2510 26100 16400 30400 24300 41200 32200 85600 244000 11400 3272000

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

DateUKEUATBEDEDKESFRITNLPLPTNOCHUSUS-NJUS-NYUS-WAAustraliaIranMalaysia
2020-04-14 93873 755290 14226 31119 131359 6511 172541 130253 162488 27419 7202 17448 6623 25936 607670 68824 202208 10694 6415 74877 4987
2020-04-15 98000 783000 14500 32400 135000 6620 177000 137000 166000 28500 7490 18100 6680 26200 631000 71500 210000 11000 6470 76600 5080
2020-04-16 102000 801000 14700 33500 138000 6750 181000 139000 169000 29400 7710 18700 6740 26400 651000 74600 216000 11200 6530 78200 5190
2020-04-17 105000 814000 14800 34500 141000 6860 185000 141000 172000 30200 7890 19200 6790 26600 669000 77100 222000 11500 6580 79800 5310
2020-04-18 108000 828000 15000 35300 144000 6990 189000 143000 175000 31000 8120 19600 6840 26800 689000 80000 227000 11800 6640 81400 5410
2020-04-19 110000 842000 15100 35900 147000 7140 193000 146000 178000 31800 8360 20000 6890 26900 707000 82900 231000 12000 6680 83000 5510
2020-04-20 113000 851000 15200 36400 149000 7310 197000 147000 180000 32400 8560 20300 6940 27100 723000 85300 234000 12300 6730 84400 5580
2020-04-21 115000 859000 15300 37000 152000 7440 201000 149000 182000 33000 8700 20500 6990 27300 739000 88000 238000 12500 6760 85800 5660
2020-04-22 117000 868000 15400 37500 153000 7590 205000 152000 183000 33600 8830 20700 7040 27500 750000 90600 241000 12800 6810 87200 5740
2020-04-23 118000 875000 15500 37900 154000 7740 208000 154000 185000 34200 9050 21000 7090 27700 762000 93000 243000 13000 6860 88500 5820

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

UKEUATBEBSCZDEDKESFIFRGRIEITNLPLPTROSENOCH
Peak date --04-0403-2603-2903-2503-2803-2804-0703-2704-0604-0404-02 --03-21 --04-0704-1004-0904-0803-2703-23
Peak daily increment 32665 875 1458 287 284 6006 332 8157 175 10266 98 5703 405 850 374 565 297 1205
Days from 100 to peak 41 18 22 11 15 27 28 25 24 34 22 27 23 28 26 32 21 18
Days from peak/2 to peak 25 13 14 11 14 15 28 14 24 19 25 17 21 24 23 29 19 12
Days since peak 10 19 16 20 17 17 7 18 8 10 12 24 7 4 5 6 18 22

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

AustraliaBrazilCanadaIranMalaysiaPhilippinesSouth AfricaUSUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date03-2804-0804-0503-3104-0203-3103-2604-0804-0504-0304-0904-0804-0304-07 --04-04 --04-0304-0504-0804-0204-10
Peak daily increment 411 1666 1502 3106 181 363 190 31831 1326 390 959 143 1179 953 1755 1651 3778 9697 531 120574
Days from 100 to peak 18 25 25 35 24 17 8 36 28 19 21 17 19 22 19 16 21 31 26 79
Days from peak/2 to peak 14 18 17 28 25 13 8 19 19 16 16 19 13 17 13 12 14 19 22 64
Days since peak 17 6 9 14 12 14 19 6 9 11 5 6 11 7 10 11 9 6 12 4

Initial visual evaluation of forecasts of Confirmed