COVID-19 short-term forecasts Deaths 2020-04-24


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 The documentation that is provided is still in progress and not peer reviewed.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.
[2020-04-16] Added scenario forecasts to all graphs now. This would now be the preferred forecast for most.
This is the first time with a peak in confirmed UK cases (also for deaths, but this is uncertain because it is at the same date).
[2020-04-17] Bird and Nielsen look into nowcasting death counts in England.
[2020-04-24] A summary of our work on short-term COVID-19 forecasting appeared as a voxeu.

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. Also available as Nuffield Economics Discussion Paper 2020-W06. Still preliminary is the documentation of the medium term forecasts.

Deaths count average forecast Europe (bold black line in graphs) 2020-04-25 to 2020-05-01

DateUKEUATBECZDEDKESFRHUIEITNLPLPTROSECH
2020-04-24 19506 94665 530 6679 214 5760 403 22524 22245 262 1014 25969 4289 494 854 567 2152 1589
2020-04-25 20200 97000 540 6870 220 5980 410 22900 22700 280 1130 26400 4410 530 880 590 2300 1630
2020-04-26 20900 99000 560 7050 220 6170 420 23300 23100 300 1230 26800 4520 560 910 620 2470 1660
2020-04-27 21600 101000 570 7220 230 6360 430 23600 23500 310 1330 27300 4640 580 940 650 2640 1700
2020-04-28 22300 103000 590 7400 240 6570 440 24000 23900 330 1440 27700 4770 620 960 680 2820 1730
2020-04-29 23100 105000 600 7570 240 6780 450 24400 24400 350 1560 28200 4890 650 990 710 3020 1770
2020-04-30 23900 108000 620 7760 250 6990 460 24700 24800 370 1700 28600 5030 680 1020 740 3240 1810
2020-05-01 24700 110000 640 7950 250 7220 470 25100 25200 390 1860 29100 5160 720 1050 780 3460 1850

Deaths count average forecast (bold black line in graphs) 2020-04-25 to 2020-05-01

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-24 3704 2384 780 689 5574 1069 477 2600 51949 1619 672 1764 1045 889 1804 1601 2556 3084 5617 16162 731 256067
2020-04-25 4020 2550 830 710 5660 1180 490 2720 54200 1720 740 1880 1110 920 1920 1660 2750 3220 5890 16600 750 265000
2020-04-26 4270 2740 890 730 5740 1290 500 2860 56200 1830 790 1990 1180 960 2040 1710 2950 3350 6160 17000 760 274000
2020-04-27 4540 2940 940 750 5820 1400 510 3010 58300 1940 840 2120 1250 1000 2170 1760 3160 3480 6440 17300 780 284000
2020-04-28 4830 3160 1000 770 5900 1520 520 3160 60400 2060 900 2250 1320 1040 2310 1810 3390 3630 6730 17700 800 293000
2020-04-29 5130 3390 1070 790 5990 1660 530 3320 62600 2180 970 2390 1400 1080 2450 1860 3640 3770 7030 18100 820 303000
2020-04-30 5460 3650 1140 810 6070 1810 540 3490 65000 2320 1040 2550 1490 1130 2610 1920 3900 3930 7350 18500 840 312000
2020-05-01 5810 3930 1210 830 6160 1970 560 3670 67400 2460 1110 2720 1580 1170 2770 1980 4180 4100 7690 18900 860 322000

Deaths count forecast Europe (bold red line in graphs) 2020-04-25 to 2020-05-01

DateUKEUATBECZDEDKESFRHUIEITNLPLPTROSECH
2020-04-24 19506 94665 530 6679 214 5760 403 22524 22245 262 1014 25969 4289 494 854 567 2152 1589
2020-04-25 20100 97000 540 6840 220 5930 410 22800 22600 280 1170 26400 4400 530 880 600 2340 1620
2020-04-26 20800 99000 550 7000 220 6110 420 23200 23000 310 1370 26800 4500 570 910 620 2540 1650
2020-04-27 21400 101000 560 7160 230 6320 430 23500 23400 330 1580 27200 4610 610 930 650 2750 1690
2020-04-28 22100 103000 570 7310 230 6520 440 23900 23700 360 1820 27600 4720 660 960 680 2980 1720
2020-04-29 22700 105000 580 7470 240 6730 450 24300 24100 380 2100 28000 4830 710 990 710 3230 1750
2020-04-30 23500 107000 590 7630 240 6940 460 24600 24500 410 2430 28400 4940 760 1020 740 3490 1790
2020-05-01 24200 109000 600 7790 250 7160 470 25000 24800 450 2800 28900 5060 820 1040 780 3780 1830

Deaths count forecast (bold red line in graphs) 2020-04-25 to 2020-05-01

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-24 3704 2384 780 689 5574 1069 477 2600 51949 1619 672 1764 1045 889 1804 1601 2556 3084 5617 16162 731 256067
2020-04-25 4010 2520 830 710 5660 1180 490 2720 53900 1710 780 1850 1140 920 1910 1650 2730 3200 5870 16600 750 264000
2020-04-26 4330 2680 890 730 5740 1260 500 2840 55700 1820 900 1950 1230 950 2020 1700 2910 3330 6130 16900 760 272000
2020-04-27 4690 2840 950 750 5830 1360 510 2960 57600 1920 1030 2070 1330 980 2140 1760 3090 3450 6390 17300 780 280000
2020-04-28 5060 3020 1010 780 5910 1460 520 3080 59400 2040 1180 2190 1440 1010 2260 1810 3290 3580 6660 17700 790 288000
2020-04-29 5470 3200 1080 800 6000 1570 530 3220 61200 2160 1340 2310 1550 1040 2390 1860 3500 3720 6940 18100 810 296000
2020-04-30 5910 3400 1150 830 6080 1690 550 3350 63100 2280 1530 2450 1680 1070 2530 1920 3730 3860 7230 18500 820 303000
2020-05-01 6380 3620 1230 860 6170 1820 560 3500 65100 2420 1750 2590 1810 1110 2680 1980 3970 4010 7540 18900 840 311000

Deaths count scenario forecast (bold green line in graphs) 2020-04-25 to 2020-05-03

DateUKEUATBECZDEDKESFRHUIEITNLPLPTROSECHBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-24 19506 94665 530 6679 214 5760 403 22524 22245 262 1014 25969 4289 494 854 567 2152 1589 3704 2384 780 689 5574 1069 477 2600 51949 1619 672 1764 1045 889 1804 1601 2556 3084 5617 16162 731 256067
2020-04-25 19900 97000 540 6890 220 5960 410 22800 22600 270 960 26400 4390 510 880 580 2250 1620 3820 2510 800 700 5650 1140 490 2710 53800 1690 700 1830 1070 920 1850 1630 2680 3190 5840 16600 750 264000
2020-04-26 20500 99000 550 7080 220 6150 420 23200 22900 280 1020 26700 4480 530 900 600 2320 1650 4010 2640 830 720 5710 1220 500 2810 55600 1760 730 1920 1110 960 1930 1680 2810 3290 6070 17000 760 272000
2020-04-27 21100 100000 560 7250 230 6340 430 23400 23300 290 1040 27100 4580 540 920 610 2400 1680 4150 2770 870 730 5760 1280 500 2910 57500 1830 750 2010 1150 990 2010 1720 2950 3380 6290 17300 770 280000
2020-04-28 21600 102000 580 7410 230 6520 440 23700 23600 300 1060 27500 4690 560 940 620 2470 1710 4280 2890 900 750 5810 1340 510 3010 59300 1890 770 2090 1190 1020 2080 1760 3090 3470 6500 17600 790 286000
2020-04-29 22100 104000 590 7560 230 6690 440 23900 23900 310 1100 27800 4780 570 960 630 2550 1740 4410 3020 930 760 5850 1450 520 3110 61100 1960 810 2180 1230 1040 2150 1800 3200 3560 6710 17900 800 294000
2020-04-30 22600 106000 600 7670 240 6860 450 24100 24200 330 1140 28200 4870 590 980 640 2620 1760 4560 3130 970 780 5880 1500 530 3190 62900 2030 840 2250 1260 1070 2230 1840 3300 3650 6890 18100 810 301000
2020-05-01 23100 107000 610 7790 240 7010 460 24300 24500 340 1170 28500 4950 610 990 650 2690 1790 4720 3240 990 790 5910 1580 540 3280 64500 2090 860 2330 1280 1100 2290 1880 3420 3740 7090 18300 820 309000
2020-05-02 23600 109000 620 7890 240 7170 460 24500 24800 350 1210 28800 5010 620 1010 650 2770 1810 4850 3350 1020 800 5950 1640 550 3360 66100 2160 880 2420 1310 1120 2350 1910 3540 3820 7270 18500 830 316000
2020-05-03 24100 110000 630 7990 240 7320 470 24600 25100 360 1240 29100 5080 630 1020 660 2840 1820 4980 3460 1050 810 5980 1710 560 3440 67400 2240 900 2480 1340 1140 2410 1940 3640 3900 7460 18700 840 322000

Peak increase in estimated trend of Deaths in Europe 2020-04-24

UKEUATBECZDEDKESFRHUIEITNLPLPTROSECH
Peak date04-1004-0704-0904-1004-0904-1504-0404-0204-07 -- --03-2804-07 --04-11 -- --04-04
Peak daily increment 886 3240 23 365 11 275 18 902 1014 826 163 33 61
Days from 100 to peak 22 34 10 17 0 23 3 20 23 24 18 13 12
Days from peak/2 to peak 16 24 18 15 14 22 14 16 17 19 18 19 17
Days since peak 14 17 15 14 15 9 20 22 17 27 17 13 20

Peak increase in estimated trend of Deaths 2020-04-24

BrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date -- -- --04-1203-19 --04-1204-2004-16 -- --04-20 --04-20 --04-14 --04-1504-1604-0904-0604-16
Peak daily increment 38 150 30 126 3071 108 43 77 143 294 779 30 11139
Days from 100 to peak 14 14 11 23 30 18 21 18 18 20 18 15 81
Days from peak/2 to peak 24 16 25 25 20 20 29 21 19 18 14 30 49
Days since peak 12 36 12 4 8 4 4 10 9 8 15 18 8

Initial visual evaluation of forecasts of Deaths