COVID-19 short-term forecasts Deaths 2020-05-13


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 The documentation that is provided is still in progress and not peer reviewed.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.
[2020-04-27] Our short-term COVID-19 forecasting paper is now available as Nuffield Economics Discussion Paper 2020-W06.
A small adjustment has been made to the scenario forecast methodology, and will be documented shortly.
[2020-04-29] See our blog entry at the International Institute of Forecasters.
US history of death counts revised in Johns Hopkins/CSSE data.
UK death counts have been revised to include the deaths in care homes. In the Johns Hopkins/CSSE data set, which we use, the entire history has been revised. So forecasts made up to 2020-04-29 cannot be compared to later outcomes. In the ECDC data set only the last observation has changed, causing a jump in the series.
[2020-05-06] The New York Times is in the process of redefining its US state data. Unfortunately, at the moment only the last observation has changed (e.g New York deaths jumped from 19645 on 2020-05-05 to 25956 a day later). This means the data is currently useless; however it does bring it close to the Johns Hopkins/CSSE count (25626 on 2020-05-06). The aggregate US count is based on JH/CSSE so unaffected. We now use Johns Hopkins/CSSE US state data, including all states with sufficient counts. So the new forecasts cannot be compared to those previously.
A minor change is that we show the graph without scenario forecast if no peak has been detected yet.
[2020-05-13] We now omit countries with fewer than 200 confirmed cases in the last week (25 for deaths).
The short-term paper has some small updates, including further comparisons with other models.
Data for Ecuador are not reliable enough for forecasting.
Switched to an improved version of scenario forecasting.

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) 2020-05-14 to 2020-05-20

DateUKEUBECZDEESFIFRHUIEITNLPLPTROSECH
2020-05-13 33186 118433 8843 290 7861 27104 284 27032 430 1497 31106 5562 861 1175 1036 3460 1870
2020-05-14 33400 119000 8900 290 7890 27200 290 27100 440 1510 31300 5590 870 1180 1050 3470 1880
2020-05-15 33700 120000 8970 300 7940 27400 290 27300 440 1530 31500 5630 880 1190 1070 3490 1880
2020-05-16 34100 121000 9040 300 8010 27600 290 27500 450 1540 31700 5680 890 1200 1090 3520 1890
2020-05-17 34400 122000 9110 300 8080 27700 290 27600 460 1550 31800 5720 910 1210 1110 3570 1900
2020-05-18 34800 123000 9180 300 8150 27900 300 27800 470 1570 32000 5760 920 1220 1130 3620 1910
2020-05-19 35200 123000 9250 310 8220 28100 300 28000 470 1580 32200 5800 940 1230 1160 3670 1920
2020-05-20 35600 124000 9330 310 8300 28200 300 28200 480 1600 32400 5840 950 1240 1180 3730 1930

Deaths count average forecast (bold black line in graphs) 2020-05-14 to 2020-05-20

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-13 13240 5425 346 509 2551 1028 6783 4220 2169 772 2212 219 3952 84119 450 594 2957 1058 3120 350 246 1827 1494 306 3790 1482 321 2315 5310 1717 4653 638 535 465 625 9714 231 324 27477 1483 4094 362 269 1172 928 974 421
2020-05-14 14000 5550 350 520 2620 1040 6830 4350 2260 780 2320 230 3980 84700 460 600 2980 1060 3140 360 250 1840 1510 310 3810 1500 320 2330 5390 1730 4690 650 540 470 640 9900 240 330 27600 1500 4120 370 270 1200 930 990 430
2020-05-15 14800 5690 360 540 2730 1060 6880 4580 2370 790 2440 230 4020 85900 470 610 3020 1070 3180 370 250 1870 1530 330 3870 1530 330 2360 5500 1770 4740 670 550 480 660 10200 240 330 27800 1510 4150 370 270 1230 940 1000 430
2020-05-16 15700 5840 370 560 2840 1070 6930 4850 2490 800 2560 240 4070 87200 490 630 3080 1070 3230 380 260 1900 1560 340 3960 1550 330 2390 5610 1810 4800 690 560 490 690 10400 250 340 28100 1530 4210 380 280 1250 960 1020 440
2020-05-17 16700 5990 370 570 2960 1090 6980 5130 2610 810 2700 250 4120 88400 510 650 3140 1080 3270 390 270 1930 1590 360 4060 1580 340 2420 5730 1850 4850 710 580 500 710 10700 260 340 28300 1560 4270 380 280 1280 980 1030 450
2020-05-18 17700 6150 380 590 3080 1110 7030 5420 2730 830 2830 260 4160 89700 530 670 3200 1100 3310 400 270 1960 1620 370 4160 1610 340 2450 5850 1900 4900 740 590 520 740 11000 270 350 28500 1590 4340 390 280 1310 1000 1050 450
2020-05-19 18800 6310 390 610 3210 1120 7080 5740 2860 840 2980 270 4210 91000 550 690 3270 1110 3360 410 280 1990 1660 390 4270 1640 350 2480 5980 1950 4960 760 610 530 760 11400 290 350 28700 1620 4410 400 290 1340 1020 1070 460
2020-05-20 20000 6470 400 630 3350 1140 7130 6070 3000 850 3130 280 4260 92400 570 710 3330 1130 3410 430 290 2030 1690 400 4380 1670 350 2510 6100 1990 5010 780 620 540 790 11700 300 360 28900 1660 4480 410 290 1380 1040 1090 470

Deaths count forecast Europe (bold red line in graphs) 2020-05-14 to 2020-05-20

DateUKEUBECZDEESFIFRHUIEITNLPLPTROSECH
2020-05-13 33186 118433 8843 290 7861 27104 284 27032 430 1497 31106 5562 861 1175 1036 3460 1870
2020-05-14 33500 119000 8890 290 7930 27300 290 27200 440 1510 31300 5610 880 1180 1050 3520 1880
2020-05-15 33900 120000 8940 300 8000 27400 290 27300 440 1520 31500 5650 890 1200 1080 3590 1890
2020-05-16 34300 121000 8990 300 8080 27600 300 27500 450 1530 31600 5700 910 1210 1100 3650 1900
2020-05-17 34700 122000 9040 300 8160 27800 300 27700 450 1540 31800 5740 930 1220 1120 3720 1910
2020-05-18 35100 123000 9080 310 8230 27900 310 27800 460 1560 32000 5780 940 1230 1150 3790 1920
2020-05-19 35500 124000 9130 310 8310 28100 310 28000 470 1570 32200 5830 960 1240 1170 3860 1930
2020-05-20 35900 125000 9180 310 8390 28300 320 28200 470 1580 32300 5870 980 1250 1200 3930 1930

Deaths count forecast (bold red line in graphs) 2020-05-14 to 2020-05-20

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-13 13240 5425 346 509 2551 1028 6783 4220 2169 772 2212 219 3952 84119 450 594 2957 1058 3120 350 246 1827 1494 306 3790 1482 321 2315 5310 1717 4653 638 535 465 625 9714 231 324 27477 1483 4094 362 269 1172 928 974 421
2020-05-14 14200 5540 360 520 2630 1040 6830 4460 2270 780 2320 230 4000 85500 470 610 3020 1070 3160 360 250 1860 1520 320 3890 1510 320 2340 5400 1760 4700 660 550 470 650 9900 240 330 27700 1520 4190 370 270 1190 950 990 430
2020-05-15 15200 5670 370 540 2730 1060 6880 4730 2390 800 2430 240 4050 86800 490 630 3080 1090 3200 370 260 1890 1540 330 4000 1540 330 2370 5510 1810 4740 680 560 480 670 10100 250 330 27900 1550 4280 380 280 1220 970 1000 430
2020-05-16 16200 5800 380 550 2840 1080 6920 5010 2500 810 2550 250 4090 88100 510 640 3140 1120 3250 380 270 1920 1560 340 4110 1570 330 2400 5620 1860 4790 700 580 490 690 10300 270 340 28100 1580 4370 380 290 1240 990 1020 440
2020-05-17 17300 5930 390 570 2940 1090 6970 5310 2620 830 2680 250 4140 89400 540 660 3210 1140 3300 400 270 1950 1580 360 4230 1600 330 2430 5730 1910 4840 720 590 490 720 10600 280 350 28300 1610 4450 390 290 1260 1010 1030 450
2020-05-18 18500 6060 400 580 3060 1110 7020 5630 2750 850 2810 260 4190 90700 560 680 3280 1160 3350 410 280 1990 1610 370 4350 1630 340 2460 5850 1960 4880 740 610 500 750 10800 290 350 28500 1650 4530 400 300 1290 1040 1040 450
2020-05-19 19700 6200 410 600 3170 1130 7060 5970 2880 860 2950 270 4240 92100 590 700 3350 1180 3400 430 290 2020 1630 390 4480 1660 340 2490 5960 2010 4930 760 630 510 780 11100 300 360 28800 1680 4610 410 310 1310 1060 1060 460
2020-05-20 21100 6340 420 620 3300 1140 7110 6320 3020 880 3100 280 4290 93400 610 720 3420 1210 3450 440 290 2050 1650 410 4610 1690 340 2520 6080 2070 4980 780 640 520 800 11400 320 370 29000 1720 4700 410 310 1340 1080 1070 460

Deaths count scenario forecast (bold purple line in graphs) 2020-05-14 to 2020-05-22

DateUKEUBECZDEESFIFRHUIEITNLPLPTROSECHBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-05-13 33186 118433 8843 290 7861 27104 284 27032 430 1497 31106 5562 861 1175 1036 3460 1870 13240 5425 346 509 2551 1028 6783 4220 2169 772 2212 219 3952 84119 450 594 2957 1058 3120 350 246 1827 1494 306 3790 1482 321 2315 5310 1717 4653 638 535 465 625 9714 231 324 27477 1483 4094 362 269 1172 928 974 421
2020-05-14 33400 119000 8900 290 7890 27300 280 27200 440 1510 31300 5600 870 1180 1050 3460 1880 14000 5540 350 520 2600 1030 6820 4400 2270 780 2340 230 3980 84700 470 610 3020 1060 3130 350 250 1850 1510 320 3830 1500 320 2330 5390 1740 4700 650 550 470 640 9900 240 330 27700 1500 4120 370 270 1190 930 980 430
2020-05-15 33700 120000 8950 290 7930 27400 290 27300 440 1520 31400 5630 880 1190 1070 3490 1890 14700 5660 370 540 2680 1050 6860 4600 2360 790 2460 230 4020 85700 480 630 3090 1060 3160 360 250 1880 1530 330 3910 1530 330 2360 5470 1780 4750 670 560 490 660 10000 250 330 27800 1530 4170 370 270 1210 950 1000 430
2020-05-16 34000 120000 9010 290 7950 27500 290 27500 450 1520 31600 5670 890 1200 1100 3520 1900 15500 5780 380 550 2760 1050 6890 4760 2460 800 2590 240 4060 86500 500 650 3160 1070 3190 370 260 1910 1550 350 4000 1550 330 2380 5540 1820 4790 690 580 490 680 10100 250 330 28000 1560 4210 380 270 1230 970 1010 430
2020-05-17 34200 121000 9050 300 8000 27700 290 27600 450 1530 31700 5700 900 1210 1120 3550 1900 16200 5890 390 570 2860 1060 6930 4900 2560 810 2720 240 4100 87400 520 660 3240 1080 3220 380 260 1940 1560 360 4110 1570 330 2400 5610 1850 4830 700 590 500 710 10200 260 340 28100 1600 4270 380 280 1250 980 1020 440
2020-05-18 34500 121000 9100 300 8020 27800 290 27800 460 1540 31800 5730 900 1220 1140 3580 1910 17000 6010 400 580 2950 1070 6960 5030 2670 820 2860 250 4140 88200 540 670 3310 1090 3250 380 270 1970 1580 370 4200 1590 330 2410 5670 1880 4860 720 600 520 730 10400 270 340 28300 1620 4350 390 280 1260 1000 1020 440
2020-05-19 34700 122000 9150 300 8040 27900 300 27900 460 1550 31900 5770 910 1230 1160 3600 1910 17700 6110 420 600 3030 1080 6990 5120 2770 830 3010 260 4170 89100 550 700 3390 1100 3280 390 270 1990 1590 390 4300 1610 340 2430 5740 1910 4890 730 620 530 750 10500 280 350 28400 1630 4410 390 280 1270 1020 1030 440
2020-05-20 34900 122000 9190 300 8040 28100 300 28000 470 1550 32100 5800 920 1240 1180 3620 1920 18400 6210 430 610 3090 1080 7020 5250 2870 840 3150 270 4200 89800 560 720 3490 1110 3320 400 280 2020 1600 400 4390 1630 340 2450 5800 1940 4920 740 630 540 760 10500 280 350 28500 1660 4450 390 280 1290 1030 1030 450
2020-05-21 35200 123000 9230 300 8060 28200 300 28100 470 1560 32200 5830 930 1240 1200 3650 1920 19000 6310 440 620 3150 1090 7050 5370 2960 850 3300 270 4220 90600 580 740 3580 1120 3340 400 280 2040 1620 410 4480 1640 340 2460 5860 1970 4940 750 640 540 780 10600 290 350 28600 1680 4510 390 280 1300 1050 1030 450
2020-05-22 35400 123000 9280 300 8100 28300 300 28200 470 1570 32300 5860 930 1250 1230 3670 1930 19600 6400 460 630 3200 1090 7090 5500 3050 860 3450 280 4240 91500 600 760 3670 1120 3360 410 280 2050 1630 420 4560 1660 340 2470 5910 2000 4970 770 650 550 800 10700 300 350 28700 1700 4510 400 280 1310 1060 1040 450

Peak increase in estimated trend of Deaths in Europe 2020-05-13

UKEUBECZDEESFIFRHUIEITNLPLPTROSECH
Peak date04-0904-0704-1004-0904-1504-0204-2104-0704-1804-2403-2804-0704-2404-1104-2904-2204-04
Peak daily increment 1014 3241 364 11 280 902 19 1014 14 110 826 163 29 33 26 117 61
Days from 100 to peak 22 34 17 0 23 20 1 23 5 22 24 18 18 13 27 25 12
Days from peak/2 to peak 16 24 15 14 22 16 23 17 25 25 19 18 28 19 35 27 17
Days since peak 34 36 33 34 28 41 22 36 25 19 46 36 19 32 14 21 39

Peak increase in estimated trend of Deaths 2020-05-13

BrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
Peak date --05-05 --05-0505-0404-1403-19 -- --04-12 --05-0804-1904-15 --05-0804-2304-2404-2004-2504-2805-0604-07 --05-0604-2904-1604-1504-2504-2904-2205-0605-0805-06 --04-30 --04-1204-0704-2905-0504-2904-0304-3004-2204-0604-05
Peak daily increment 177 18 118 39 150 28 10 125 2279 38 89 57 120 13 9 55 46 125 50 13 75 265 65 148 26 20 23 317 11 1066 61 177 23 10 39 58 32 13
Days from 100 to peak 34 23 29 16 14 11 9 22 29 26 26 20 18 5 4 35 7 36 25 1 19 24 22 25 19 27 23 29 0 18 25 32 14 -8 26 7 14 -3
Days from peak/2 to peak 35 37 32 26 16 25 37 24 21 36 30 27 20 26 29 41 11 37 32 20 22 23 26 26 33 41 38 29 17 13 31 33 33 5 35 13 29 11
Days since peak 8 8 9 29 55 31 5 24 28 5 20 19 23 18 15 7 36 7 14 27 28 18 14 21 7 5 7 13 31 36 14 8 14 40 13 21 37 38

Initial visual evaluation of forecasts of Deaths