COVID-19 short-term forecasts Deaths 2020-04-22


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 The documentation that is provided is still in progress and not peer reviewed.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

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) 2020-04-23 to 2020-04-29

DateUKEUATBECZDEDKESFRHUIEITNLPLPTROSECH
2020-04-22 18100 90154 510 6262 208 5279 384 21717 21340 225 769 25085 4054 426 785 524 1937 1509
2020-04-23 18800 92000 530 6470 210 5490 390 22100 21900 240 820 25500 4180 450 810 550 2080 1550
2020-04-24 19500 94000 540 6680 220 5670 400 22500 22400 250 880 26000 4300 470 830 570 2210 1580
2020-04-25 20300 97000 560 6900 230 5860 410 22800 22900 270 940 26400 4420 500 860 590 2350 1620
2020-04-26 21000 99000 570 7130 230 6050 420 23200 23400 290 1010 26900 4550 520 890 620 2490 1660
2020-04-27 21900 101000 590 7370 240 6250 430 23600 23900 310 1080 27400 4680 550 910 640 2650 1700
2020-04-28 22700 103000 600 7610 240 6460 440 24000 24500 330 1160 27900 4820 580 940 670 2820 1740
2020-04-29 23600 106000 620 7870 250 6670 450 24400 25100 350 1240 28400 4960 610 970 690 3000 1780

Deaths count average forecast (bold black line in graphs) 2020-04-23 to 2020-04-29

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-22 2906 2075 681 635 5391 857 446 2376 46583 1425 526 1544 926 837 1577 1473 2182 2812 5063 15302 696 236401
2020-04-23 3060 2250 720 650 5480 940 460 2510 48800 1520 560 1670 980 890 1680 1530 2370 2940 5340 15700 710 246000
2020-04-24 3210 2420 770 680 5550 1020 470 2670 51200 1610 590 1800 1040 940 1780 1570 2560 3050 5590 16200 730 257000
2020-04-25 3370 2610 810 700 5630 1110 480 2830 53800 1720 620 1930 1100 990 1880 1620 2770 3160 5850 16600 750 268000
2020-04-26 3530 2820 850 720 5710 1210 490 3000 56500 1830 650 2070 1160 1050 2000 1670 2990 3280 6120 17000 770 279000
2020-04-27 3700 3040 900 740 5790 1320 500 3190 59400 1940 680 2220 1230 1110 2120 1710 3230 3400 6400 17500 780 290000
2020-04-28 3890 3280 950 760 5870 1450 520 3380 62500 2070 720 2380 1300 1180 2240 1770 3490 3520 6700 17900 800 302000
2020-04-29 4080 3550 1010 790 5950 1590 530 3590 65800 2200 760 2560 1380 1250 2380 1820 3770 3660 7020 18400 820 314000

Deaths count forecast Europe (bold red line in graphs) 2020-04-23 to 2020-04-29

DateUKEUATBECZDEDKESFRHUIEITNLPLPTROSECH
2020-04-22 18100 90154 510 6262 208 5279 384 21717 21340 225 769 25085 4054 426 785 524 1937 1509
2020-04-23 18700 92000 530 6440 210 5450 390 22000 21800 240 820 25500 4160 460 810 540 2040 1540
2020-04-24 19400 94000 540 6630 220 5640 400 22300 22200 250 870 25900 4270 490 830 570 2190 1570
2020-04-25 20100 96000 560 6810 220 5830 410 22700 22700 260 920 26300 4390 520 850 590 2350 1610
2020-04-26 20800 98000 570 7000 230 6030 420 23000 23100 280 980 26800 4510 560 870 610 2510 1640
2020-04-27 21600 100000 590 7180 240 6230 430 23400 23500 290 1030 27200 4630 600 900 640 2690 1670
2020-04-28 22300 103000 610 7370 240 6440 440 23700 24000 310 1090 27600 4760 640 920 660 2890 1710
2020-04-29 23100 105000 630 7560 250 6660 450 24100 24400 330 1160 28100 4890 690 950 690 3090 1740

Deaths count forecast (bold red line in graphs) 2020-04-23 to 2020-04-29

DateBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-22 2906 2075 681 635 5391 857 446 2376 46583 1425 526 1544 926 837 1577 1473 2182 2812 5063 15302 696 236401
2020-04-23 3050 2170 720 650 5470 920 460 2510 48400 1480 560 1670 990 870 1660 1520 2390 2910 5300 15700 710 245000
2020-04-24 3200 2300 750 670 5550 1010 470 2640 50200 1540 600 1810 1060 910 1750 1570 2620 3020 5560 16100 730 254000
2020-04-25 3340 2440 800 690 5630 1110 480 2790 51900 1600 640 1960 1130 950 1850 1620 2880 3140 5840 16400 740 263000
2020-04-26 3480 2590 840 700 5710 1220 480 2950 53700 1670 680 2120 1200 990 1950 1670 3150 3260 6120 16800 760 272000
2020-04-27 3630 2750 880 720 5790 1340 490 3110 55500 1750 730 2290 1280 1030 2060 1720 3450 3380 6430 17200 780 282000
2020-04-28 3780 2920 930 740 5870 1470 500 3280 57400 1820 780 2470 1370 1080 2170 1780 3780 3510 6740 17600 790 291000
2020-04-29 3940 3100 980 750 5960 1620 510 3470 59300 1900 830 2670 1460 1130 2290 1840 4150 3650 7080 18000 810 300000

Deaths count scenario forecast (bold green line in graphs) 2020-04-23 to 2020-05-01

DateUKEUATBECZDEDKESFRHUIEITNLPLPTROSECHBrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
2020-04-22 18100 90154 510 6262 208 5279 384 21717 21340 225 769 25085 4054 426 785 524 1937 1509 2906 2075 681 635 5391 857 446 2376 46583 1425 526 1544 926 837 1577 1473 2182 2812 5063 15302 696 236401
2020-04-23 18600 92000 520 6480 210 5440 390 22000 21900 240 820 25500 4140 440 810 540 1970 1540 3010 2180 710 660 5460 890 460 2490 48000 1460 540 1620 960 870 1630 1500 2280 2900 5220 15800 710 244000
2020-04-24 19200 94000 530 6690 220 5640 400 22200 22500 250 860 25900 4240 450 830 550 2040 1570 3110 2320 740 680 5530 950 470 2600 49700 1520 560 1700 1000 910 1700 1540 2410 3000 5440 16200 720 250000
2020-04-25 19800 96000 540 6890 220 5840 410 22500 23100 260 900 26300 4340 470 850 560 2120 1600 3230 2450 770 700 5580 1010 490 2710 51100 1590 580 1780 1030 940 1770 1590 2550 3090 5660 16700 730 258000
2020-04-26 20300 98000 540 7100 230 6020 420 22700 23600 270 950 26600 4430 480 870 580 2190 1620 3350 2570 800 720 5630 1070 500 2810 51800 1650 590 1850 1070 980 1840 1630 2670 3180 5860 17100 740 265000
2020-04-27 20900 100000 550 7300 230 6190 420 22900 24100 290 990 27000 4540 490 890 590 2260 1650 3460 2690 820 740 5680 1130 520 2910 53300 1710 610 1930 1110 1010 1910 1670 2800 3270 6060 17400 750 273000
2020-04-28 21400 101000 560 7480 240 6350 430 23100 24500 300 1030 27300 4620 510 910 610 2340 1670 3570 2800 850 760 5710 1180 530 3010 54000 1760 620 1990 1140 1040 1970 1700 2930 3350 6250 17700 760 279000
2020-04-29 21900 103000 560 7660 240 6510 440 23300 24900 310 1070 27600 4710 520 930 620 2410 1690 3670 2910 870 780 5750 1230 540 3100 55700 1820 630 2070 1170 1070 2030 1740 3050 3440 6440 18100 770 285000
2020-04-30 22400 105000 570 7830 240 6670 450 23500 25300 320 1100 27900 4790 530 940 630 2470 1720 3770 3020 890 800 5800 1270 540 3190 56100 1870 650 2150 1200 1100 2080 1770 3170 3520 6630 18400 780 291000
2020-05-01 22800 106000 570 7970 250 6820 450 23700 25600 330 1130 28200 4870 540 950 640 2530 1730 3880 3110 920 810 5840 1290 550 3270 57400 1930 660 2220 1230 1130 2140 1800 3280 3590 6820 18600 790 298000

Peak increase in estimated trend of Deaths in Europe 2020-04-22

UKEUATBECZDEDKESFRHUIEITNLPLPTROSECH
Peak date04-1004-0704-0804-1004-0904-1504-0404-0204-0704-18 --03-2804-07 --04-1104-09 --04-04
Peak daily increment 880 3234 23 361 11 271 18 902 1022 14 826 165 33 24 61
Days from 100 to peak 22 34 9 17 0 23 3 20 23 5 24 18 13 7 12
Days from peak/2 to peak 16 24 17 15 14 22 14 16 17 25 19 18 19 15 17
Days since peak 12 15 14 12 13 7 18 20 15 4 25 15 11 13 18

Peak increase in estimated trend of Deaths 2020-04-22

BrazilCanadaIndiaIndonesiaIranMexicoPhilippinesTurkeyUSUS-CAUS-COUS-CTUS-FLUS-GAUS-ILUS-LAUS-MAUS-MIUS-NJUS-NYUS-WAWorld
Peak date04-18 -- --04-1203-19 --04-12 --04-16 -- -- -- -- -- --04-14 --04-1504-1504-1104-0604-16
Peak daily increment 199 37 150 28 3089 77 143 296 777 30 11265
Days from 100 to peak 21 14 14 11 30 18 18 19 20 15 81
Days from peak/2 to peak 21 24 16 25 20 21 19 17 16 30 48
Days since peak 4 10 34 10 6 8 7 7 11 16 6

Initial visual evaluation of forecasts of Deaths