COVID-19 short-term forecasts Deaths 2020-05-15


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 The documentation that is provided is still in progress and not peer reviewed.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) 2020-05-16 to 2020-05-22

DateUKEUBEDEESFIFRHUIEITNLPLPTROSECH
2020-05-15 33998 120329 8959 7897 27459 293 27485 442 1518 31610 5643 907 1190 1070 3646 1878
2020-05-16 34300 121000 9010 7940 27600 300 27600 450 1530 31700 5670 920 1200 1090 3670 1880
2020-05-17 34700 122000 9070 8000 27800 300 27800 450 1540 31900 5710 930 1210 1110 3720 1890
2020-05-18 35100 123000 9130 8060 27900 300 28000 460 1550 32100 5750 950 1220 1130 3770 1900
2020-05-19 35500 124000 9180 8120 28100 310 28200 460 1560 32300 5790 970 1230 1150 3820 1910
2020-05-20 35800 124000 9240 8180 28200 310 28400 470 1570 32500 5830 990 1240 1170 3870 1920
2020-05-21 36300 125000 9300 8250 28400 320 28600 470 1590 32700 5870 1010 1250 1200 3930 1930
2020-05-22 36700 126000 9360 8310 28600 320 28800 480 1600 32900 5910 1030 1260 1220 4000 1940

Deaths count average forecast (bold black line in graphs) 2020-05-16 to 2020-05-22

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
2020-05-15 14962 5679 394 546 2753 1076 6902 4767 2392 806 2418 247 4055 87530 483 650 3136 1149 3282 368 270 1917 1563 336 4056 1550 332 2382 5588 1820 4765 683 563 493 667 10148 253 343 27878 1581 4422 380 285 1280 978 991 445
2020-05-16 15700 5800 400 550 2830 1080 6940 5030 2500 820 2500 250 4090 88300 500 680 3170 1160 3310 380 280 1930 1580 350 4210 1560 340 2400 5650 1850 4810 700 570 500 690 10400 260 340 28100 1630 4470 380 290 1310 990 1000 450
2020-05-17 16600 5940 410 570 2940 1090 6990 5330 2620 830 2590 260 4140 89700 520 710 3240 1170 3340 380 290 1960 1610 360 4390 1590 340 2430 5730 1900 4860 720 590 510 710 10600 270 350 28300 1690 4520 390 290 1360 1020 1020 450
2020-05-18 17600 6070 430 590 3050 1110 7030 5660 2750 850 2700 270 4190 91000 540 740 3320 1180 3380 400 300 1990 1650 380 4570 1620 340 2460 5820 1950 4910 740 600 520 730 10900 290 350 28500 1750 4600 400 300 1410 1040 1030 460
2020-05-19 18700 6220 450 600 3170 1130 7080 6020 2890 860 2800 280 4240 92400 570 780 3390 1200 3420 410 310 2020 1680 400 4760 1650 350 2490 5920 2000 4960 760 610 530 760 11200 300 360 28700 1820 4670 410 310 1460 1060 1050 470
2020-05-20 19800 6360 470 620 3290 1150 7130 6400 3030 880 2920 290 4280 93700 590 810 3460 1220 3470 420 320 2060 1720 410 4960 1680 350 2520 6030 2050 5020 790 630 540 780 11500 310 360 28900 1890 4740 420 320 1510 1090 1060 470
2020-05-21 21000 6510 480 640 3410 1160 7180 6800 3180 900 3030 300 4330 95200 610 850 3540 1240 3520 430 330 2090 1760 430 5170 1710 360 2550 6140 2110 5070 810 640 550 810 11800 330 370 29100 1970 4810 420 320 1560 1120 1080 480
2020-05-22 22300 6660 510 660 3540 1180 7230 7230 3340 910 3160 310 4380 96600 640 890 3620 1260 3570 440 350 2120 1790 450 5390 1740 360 2590 6260 2160 5120 830 660 560 840 12200 340 380 29400 2040 4890 430 330 1620 1140 1090 490

Deaths count forecast Europe (bold red line in graphs) 2020-05-16 to 2020-05-22

DateUKEUBEDEESFIFRHUIEITNLPLPTROSECH
2020-05-15 33998 120329 8959 7897 27459 293 27485 442 1518 31610 5643 907 1190 1070 3646 1878
2020-05-16 34400 121000 9020 7940 27600 300 27700 450 1530 31800 5680 930 1200 1090 3720 1890
2020-05-17 34700 122000 9080 7990 27800 300 27900 450 1540 32000 5720 950 1210 1110 3800 1900
2020-05-18 35100 123000 9130 8030 27900 310 28100 460 1550 32200 5760 970 1220 1130 3870 1900
2020-05-19 35500 124000 9190 8070 28100 310 28200 460 1560 32400 5810 990 1220 1150 3950 1910
2020-05-20 35800 124000 9250 8110 28200 320 28400 470 1580 32600 5850 1010 1230 1170 4020 1920
2020-05-21 36200 125000 9300 8150 28400 320 28600 470 1590 32800 5890 1030 1240 1190 4100 1920
2020-05-22 36600 126000 9360 8190 28600 320 28800 480 1600 33000 5930 1050 1250 1210 4180 1930

Deaths count forecast (bold red line in graphs) 2020-05-16 to 2020-05-22

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
2020-05-15 14962 5679 394 546 2753 1076 6902 4767 2392 806 2418 247 4055 87530 483 650 3136 1149 3282 368 270 1917 1563 336 4056 1550 332 2382 5588 1820 4765 683 563 493 667 10148 253 343 27878 1581 4422 380 285 1280 978 991 445
2020-05-16 15900 5800 420 560 2860 1090 6950 5100 2550 820 2500 260 4100 89000 510 700 3210 1180 3340 380 280 1950 1590 350 4280 1580 340 2410 5680 1870 4810 700 580 500 690 10300 260 350 28100 1690 4530 390 290 1320 1000 1000 450
2020-05-17 16900 5910 440 580 2970 1110 7000 5430 2710 840 2600 270 4150 90300 530 750 3290 1210 3400 390 290 1980 1620 360 4510 1610 340 2440 5790 1920 4860 720 590 510 710 10600 280 350 28300 1780 4640 390 300 1370 1030 1010 460
2020-05-18 17900 6030 470 590 3080 1130 7040 5780 2880 850 2700 280 4200 91700 560 810 3360 1240 3450 400 310 2020 1640 380 4740 1640 340 2470 5900 1970 4900 740 600 530 740 10800 290 360 28500 1880 4740 400 310 1420 1050 1020 470
2020-05-19 19000 6160 490 610 3190 1150 7090 6150 3050 870 2800 290 4250 93100 590 860 3440 1270 3510 410 320 2050 1670 390 4990 1660 350 2500 6010 2020 4950 760 610 540 770 11100 300 370 28700 1970 4850 410 320 1470 1070 1030 470
2020-05-20 20100 6280 520 630 3300 1170 7140 6540 3230 890 2900 300 4300 94400 620 920 3520 1300 3570 420 330 2080 1700 410 5240 1690 350 2530 6120 2070 5000 780 630 550 790 11300 310 380 28900 2080 4950 420 320 1520 1100 1040 480
2020-05-21 21300 6410 550 640 3430 1190 7180 6950 3430 900 3010 320 4340 95800 650 980 3600 1340 3620 440 350 2110 1730 420 5510 1720 360 2560 6230 2120 5040 800 640 560 820 11700 330 390 29100 2190 5060 430 330 1580 1130 1050 490
2020-05-22 22600 6540 580 660 3550 1210 7230 7390 3640 920 3130 330 4390 97300 680 1050 3680 1370 3680 450 360 2150 1760 440 5790 1750 360 2590 6350 2180 5090 820 650 570 850 12000 340 390 29300 2310 5170 440 340 1640 1150 1060 500

Deaths count scenario forecast (bold purple line in graphs) 2020-05-16 to 2020-05-24

DateUKEUBEDEESFIFRHUIEITNLPLPTROSECHBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth 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
2020-05-15 33998 120329 8959 7897 27459 293 27485 442 1518 31610 5643 907 1190 1070 3646 1878 14962 5679 394 546 2753 1076 6902 4767 2392 806 2418 247 4055 87530 483 650 3136 1149 3282 368 270 1917 1563 336 4056 1550 332 2382 5588 1820 4765 683 563 493 667 10148 253 343 27878 1581 4422 380 285 1280 978 991 445
2020-05-16 34200 121000 9020 7960 27600 300 27600 450 1530 31700 5670 910 1200 1090 3650 1890 15800 5800 410 560 2830 1080 6940 5030 2500 820 2510 260 4100 88000 510 690 3190 1150 3300 380 280 1930 1570 350 4200 1560 330 2400 5660 1860 4810 700 570 500 690 10200 270 340 28000 1620 4430 390 300 1320 1000 1000 450
2020-05-17 34500 122000 9070 7990 27800 300 27800 450 1530 31900 5700 920 1210 1110 3690 1890 16700 5930 430 570 2930 1090 6980 5300 2610 830 2620 270 4140 89100 530 730 3250 1180 3350 380 290 1960 1590 370 4340 1590 340 2420 5760 1900 4850 720 580 500 710 10300 280 350 28200 1670 4520 390 310 1360 1020 1010 450
2020-05-18 34800 122000 9120 8040 27900 310 27900 460 1540 32000 5740 940 1220 1120 3730 1900 17500 6040 450 590 3010 1100 7020 5550 2720 840 2720 280 4170 90200 540 760 3310 1200 3390 390 300 1980 1610 380 4490 1610 340 2440 5840 1940 4880 730 600 510 730 10500 290 350 28300 1710 4570 400 320 1400 1040 1010 450
2020-05-19 35000 123000 9170 8080 28000 310 28000 460 1550 32200 5770 950 1230 1140 3760 1900 18300 6160 470 600 3100 1110 7060 5790 2860 850 2830 290 4200 91000 560 780 3370 1220 3420 400 310 2000 1620 400 4650 1630 340 2460 5930 1980 4910 750 610 520 760 10600 310 360 28500 1760 4640 400 330 1450 1060 1020 450
2020-05-20 35200 123000 9210 8130 28200 310 28200 470 1560 32300 5800 960 1230 1160 3800 1910 19100 6270 490 610 3180 1120 7100 6040 2990 860 2910 300 4230 92000 580 800 3420 1240 3470 410 330 2020 1640 410 4800 1640 350 2480 6000 2010 4940 760 620 530 780 10700 320 360 28600 1820 4700 410 350 1510 1080 1020 450
2020-05-21 35500 124000 9250 8170 28300 310 28300 470 1570 32400 5840 970 1240 1170 3830 1910 19800 6370 510 630 3260 1120 7140 6250 3090 870 3010 310 4260 92900 590 830 3470 1240 3500 410 340 2050 1650 420 4980 1670 350 2490 6070 2050 4970 780 630 530 800 10700 330 360 28700 1860 4740 410 360 1560 1100 1020 460
2020-05-22 35700 124000 9290 8210 28400 320 28300 480 1570 32600 5870 970 1250 1180 3860 1920 20600 6480 530 640 3340 1130 7170 6490 3230 870 3100 330 4280 93700 600 850 3520 1250 3530 420 350 2070 1670 430 5110 1680 350 2510 6150 2080 4990 790 630 540 830 10800 350 370 28800 1920 4800 410 370 1610 1110 1030 460
2020-05-23 35900 125000 9330 8260 28500 320 28400 480 1580 32700 5890 980 1250 1200 3900 1920 21200 6580 550 650 3420 1140 7200 6700 3340 880 3200 340 4310 94300 620 860 3570 1260 3550 420 360 2100 1680 450 5210 1690 360 2520 6230 2110 5020 800 640 550 850 10900 360 370 28900 1960 4860 420 380 1650 1130 1030 460
2020-05-24 36100 125000 9370 8300 28600 320 28500 490 1580 32800 5920 990 1260 1210 3930 1920 21800 6670 560 660 3500 1150 7240 6910 3460 890 3290 350 4330 95100 630 880 3610 1270 3580 430 380 2120 1700 460 5320 1710 360 2540 6290 2150 5040 820 660 550 870 10900 370 370 29100 2010 4910 420 400 1690 1140 1030 460

Peak increase in estimated trend of Deaths in Europe 2020-05-15

UKEUBEDEESFIFRHUIEITNLPLPTROSECH
Peak date04-0904-0704-1004-1504-0204-2104-0704-1804-2403-2804-0704-2404-1104-2904-2204-04
Peak daily increment 1014 3241 364 280 903 19 1014 14 110 826 163 29 33 26 117 61
Days from 100 to peak 22 34 17 23 20 1 23 5 22 24 18 18 13 27 25 12
Days from peak/2 to peak 16 24 15 22 16 23 17 25 25 19 18 28 19 35 27 17
Days since peak 36 38 35 30 43 24 38 27 21 48 38 21 34 16 23 41

Peak increase in estimated trend of Deaths 2020-05-15

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 -- --04-1205-11 --04-1904-15 -- --04-2304-2404-2004-25 --05-0604-07 -- --04-2904-1604-1504-2504-2904-2205-0605-0805-06 --04-30 --04-1204-07 --05-0504-2904-03 --04-2204-0604-05
Peak daily increment 177 18 118 39 150 28 100 125 2279 89 57 120 13 54 46 50 13 75 265 64 148 25 20 23 310 11 1066 175 21 10 58 32 13
Days from 100 to peak 34 23 29 16 14 11 30 22 29 26 20 18 5 35 7 25 1 19 24 22 25 19 27 23 29 0 18 32 14 -8 7 14 -3
Days from peak/2 to peak 35 37 32 26 16 25 35 24 21 30 27 20 26 41 11 32 20 22 23 27 26 33 41 38 29 17 13 33 33 5 13 29 11
Days since peak 10 10 11 31 57 33 4 26 30 22 21 25 20 9 38 16 29 30 20 16 23 9 7 9 15 33 38 10 16 42 23 39 40

Initial visual evaluation of forecasts of Deaths