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


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 The documentation that is provided is still in progress and not peer reviewed.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

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.
  • 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) 2020-04-05 to 2020-04-11

DateUKEUATBEBSCZDEDKESFRIEITNLPLPTROSENOCH
2020-04-04 41903 534041 11781 18431 4539 4472 96092 4077 126168 89953 4604 124632 16627 3627 10524 3613 6443 5550 20505
2020-04-05 47000 584000 12100 19900 4810 4760 102000 4440 133000 104000 5010 129000 17600 4010 11300 4020 7000 5760 21400
2020-04-06 53000 642000 12600 21300 5040 5010 108000 4820 141000 117000 5410 133000 18600 4410 12300 4360 7700 5960 22300
2020-04-07 61000 705000 13000 22700 5290 5260 114000 5230 148000 132000 5840 138000 19700 4850 13300 4740 8500 6160 23200
2020-04-08 69000 774000 13400 24300 5540 5520 120000 5680 156000 148000 6300 142000 20700 5340 14300 5160 9400 6370 24200
2020-04-09 78000 849000 13900 25900 5810 5790 126000 6170 165000 168000 6810 146000 21900 5880 15500 5620 10300 6590 25200
2020-04-10 89000 930000 14300 27700 6100 6090 133000 6700 174000 192000 7360 151000 23100 6480 16800 6130 11300 6820 26300
2020-04-11 101000 1017000 14800 29700 6400 6400 141000 7280 183000 220000 7960 156000 24400 7130 18200 6680 12400 7050 27400

Confirmed count average forecast (bold black line in graphs) 2020-04-05 to 2020-04-11

DateAustraliaBrazilCanadaIranMalaysiaPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WA
2020-04-04 5550 10360 12978 55743 3483 3094 308850 13796 4574 5276 11537 6383 10357 12496 11736 14225 34124 114996 7498
2020-04-05 5770 11800 14400 58800 3700 3450 350000 15400 5000 5900 13000 7100 11800 14700 13100 16100 38800 128000 8000
2020-04-06 6020 13200 16100 62500 3950 3940 399000 17400 5420 6600 14500 8000 13400 17000 14300 18200 43400 142000 8400
2020-04-07 6270 14800 18000 66400 4210 4480 455000 19500 5860 7400 16300 9000 15200 19700 15600 20600 48600 157000 8900
2020-04-08 6530 16600 20200 70500 4490 5100 519000 22000 6340 8300 18300 10200 17300 22900 17100 23200 54500 174000 9400
2020-04-09 6790 18600 22700 74900 4780 5800 592000 24700 6860 9400 20500 11500 19700 26600 18700 26300 61100 193000 9900
2020-04-10 7080 20800 25400 79700 5090 6590 676000 27800 7440 10700 23000 13100 22400 30900 20500 29600 68600 215000 10500
2020-04-11 7360 23400 28500 84700 5430 7480 772000 31300 8050 12200 25800 14900 25500 35900 22400 33500 77100 238000 11000

Confirmed count forecast Europe (bold red line in graphs) 2020-04-05 to 2020-04-11

DateUKEUATBEBSCZDEDKESFRIEITNLPLPTROSENOCH
2020-04-04 41903 534041 11781 18431 4539 4472 96092 4077 126168 89953 4604 124632 16627 3627 10524 3613 6443 5550 20505
2020-04-05 46200 606000 12100 19700 4790 4720 101000 4410 133000 100000 5020 129000 17500 3950 11200 4110 7100 5740 21400
2020-04-06 51000 689000 12500 21100 5030 4960 107000 4780 139000 116000 5480 133000 18400 4300 11800 4660 7800 5930 22300
2020-04-07 56200 780000 12800 22400 5290 5210 112000 5170 146000 136000 5980 137000 19300 4670 12600 5280 8500 6130 23200
2020-04-08 62000 882000 13100 23800 5560 5470 117000 5600 152000 159000 6520 141000 20200 5070 13300 5960 9300 6330 24100
2020-04-09 68400 999000 13400 25200 5840 5740 123000 6060 159000 185000 7110 146000 21100 5510 14100 6740 10100 6530 25100
2020-04-10 75400 1130000 13700 26700 6130 6020 128000 6560 166000 215000 7760 150000 22000 5980 14900 7620 11000 6750 26100
2020-04-11 83200 1278000 14100 28200 6440 6310 134000 7100 174000 251000 8470 155000 23000 6500 15700 8620 12000 6970 27200

Confirmed count forecast (bold red line in graphs) 2020-04-05 to 2020-04-11

DateAustraliaBrazilCanadaIranMalaysiaPhilippinesUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WA
2020-04-04 5550 10360 12978 55743 3483 3094 308850 13796 4574 5276 11537 6383 10357 12496 11736 14225 34124 114996 7498
2020-04-05 5730 11700 14700 58600 3720 3390 350000 15300 4950 5370 12700 6800 11200 13900 12900 15800 38500 128000 7800
2020-04-06 5920 13200 16500 61500 3970 3640 398000 17000 5330 5610 13800 7400 12400 15700 14100 17600 43600 143000 8300
2020-04-07 6110 14900 18300 64600 4240 3900 454000 18900 5740 5910 15100 8000 13800 17700 15400 19700 49300 160000 8800
2020-04-08 6290 16800 20500 67700 4530 4180 519000 21100 6160 6230 16400 8600 15300 19900 16800 21800 55600 179000 9300
2020-04-09 6480 19000 22900 71000 4840 4480 594000 23500 6600 6570 17900 9300 16900 22300 18200 24300 62800 200000 9900
2020-04-10 6670 21400 25700 74400 5170 4800 681000 26300 7080 6930 19500 10100 18800 25100 19800 27000 70900 223000 10500
2020-04-11 6860 24100 28800 78100 5520 5140 781000 29400 7590 7310 21300 10900 20800 28100 21500 30100 80100 249000 11100

Confirmed count scenario forecast (bold green line in graphs) 2020-04-05 to 2020-04-13

DateUKEUATBEDEDKESFRITNLPLPTNOCHUS-WAAustraliaIranMalaysia
2020-04-04 41903 534041 11781 18431 96092 4077 126168 89953 124632 16627 3627 10524 5550 20505 7498 5550 55743 3483
2020-04-05 46100 557000 12400 19300 101000 4370 133000 83000 128000 17400 3990 11000 5720 21200 7800 5780 57900 3600
2020-04-06 50500 582000 12800 20100 107000 4610 139000 91000 131000 18100 4320 11600 5910 21900 8300 5990 59800 3710
2020-04-07 54000 605000 13200 21000 111000 4850 145000 98000 134000 18800 4620 12100 6080 22500 8700 6200 61600 3820
2020-04-08 57300 628000 13500 21900 115000 5060 151000 102000 137000 19500 4940 12400 6220 23100 9000 6410 63200 3920
2020-04-09 60000 649000 13900 22800 119000 5230 157000 103000 141000 20200 5150 12900 6390 23700 9300 6620 65000 4030
2020-04-10 62800 669000 14300 23800 123000 5410 162000 103000 144000 20800 5350 13300 6550 24200 9700 6810 66700 4130
2020-04-11 65400 689000 14700 24800 127000 5560 168000 104000 147000 21500 5520 13800 6700 24600 10000 7010 68400 4230
2020-04-12 68000 707000 15100 25900 131000 5740 173000 107000 149000 22100 5690 14200 6850 25100 10300 7190 70200 4320
2020-04-13 70100 726000 15500 26900 135000 5880 178000 110000 152000 22700 5820 14600 6990 25500 10500 7390 72000 4410

Initial visual evaluation of forecasts of Confirmed