COVID-19 short-term forecasts Deaths 2020-06-04


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 The documentation that is provided is still in progress and not peer reviewed.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. 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.
[2020-05-18] Minor fixes to the improved version of scenario forecasting, backported to 2020-05-13.
[2020-05-20] Problem with UK confirmed cases: negative daily count. This makes the forecasts temporarily unreliable.
Updated the second paper.
[2020-06-04] Data issues with confirmed cases for France.
Added an appendix to the short term paper with further forecast comparisons for European and Latin American countries.
Both Sweden and Iran have lost their peak in confirmed cases. For Sweden the previous peak was on 24 April (daily peak of 656 cases), for Iran it was on 31 March (peak of 3116). For Iran this looks like a second wave, with increasing daily counts for the last four weeks. For Sweden this is a sudden jump in confirmed cases in the last two days, compared to a fairly steady weekly pattern over the previous six weeks.

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) 2020-06-05 to 2020-06-11

DateUKEUBEDEFRHUITNLPLPTROSE
2020-06-04 39904 127415 9548 8635 29010 539 33689 5990 1117 1455 1305 4562
2020-06-05 40200 128000 9570 8650 29100 540 33800 6000 1120 1470 1310 4580
2020-06-06 40400 128000 9590 8670 29100 550 33800 6010 1130 1480 1320 4610
2020-06-07 40700 128000 9610 8690 29200 550 33900 6030 1140 1490 1330 4650
2020-06-08 41000 129000 9630 8720 29200 550 34000 6040 1150 1500 1340 4680
2020-06-09 41300 129000 9650 8740 29300 560 34100 6050 1160 1510 1350 4720
2020-06-10 41600 129000 9670 8770 29300 560 34100 6060 1170 1520 1360 4760
2020-06-11 41900 129000 9690 8790 29400 560 34200 6080 1180 1530 1370 4790

Deaths count average forecast (bold black line in graphs) 2020-06-05 to 2020-06-11

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NHUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-04 34021 7717 1356 1099 6363 1721 8071 12545 5031 984 5376 848 4630 108211 651 996 4444 1512 4007 385 2607 2118 583 5735 2052 458 2772 7190 2628 5521 1115 768 794 1006 273 11970 383 430 30174 2340 5832 397 1776 1445 1138 626
2020-06-05 35000 7770 1400 1140 6630 1740 8120 12800 5110 990 5520 870 4650 109000 660 1010 4490 1520 4020 390 2630 2140 590 5760 2060 460 2780 7250 2650 5550 1120 770 800 1010 280 12000 390 430 30200 2350 5860 400 1780 1460 1140 630
2020-06-06 36400 7840 1470 1180 6920 1770 8180 13200 5200 1000 5680 910 4680 110000 660 1020 4540 1530 4050 390 2650 2160 600 5810 2080 470 2800 7320 2670 5580 1140 780 810 1030 290 12100 390 430 30300 2380 5910 400 1800 1470 1150 640
2020-06-07 37900 7930 1550 1230 7230 1790 8240 13700 5300 1000 5860 950 4700 110000 670 1030 4590 1540 4070 390 2680 2180 610 5870 2090 470 2810 7400 2700 5610 1150 790 830 1040 300 12200 400 440 30400 2410 5960 400 1820 1480 1150 640
2020-06-08 39400 8020 1630 1280 7550 1820 8300 14400 5400 1010 6030 1000 4720 111000 680 1050 4650 1550 4090 400 2700 2210 620 5930 2110 480 2830 7470 2740 5640 1170 800 840 1060 310 12300 400 440 30500 2440 6000 410 1830 1500 1160 650
2020-06-09 41000 8110 1710 1330 7880 1850 8360 15100 5510 1020 6220 1050 4740 112000 690 1060 4710 1570 4110 400 2730 2230 630 5990 2130 480 2840 7550 2770 5670 1190 800 850 1070 320 12500 410 440 30600 2480 6060 410 1850 1510 1160 660
2020-06-10 42700 8200 1800 1380 8230 1870 8420 15800 5630 1030 6410 1100 4770 113000 690 1080 4770 1580 4140 400 2750 2260 650 6060 2150 490 2860 7630 2800 5700 1200 810 860 1090 330 12600 420 440 30700 2510 6120 420 1880 1530 1170 660
2020-06-11 44400 8300 1890 1430 8590 1900 8480 16600 5750 1030 6610 1150 4790 114000 700 1100 4830 1590 4160 410 2790 2280 660 6130 2160 490 2880 7700 2840 5730 1220 820 870 1110 340 12700 420 450 30800 2550 6180 420 1900 1540 1170 670

Deaths count forecast Europe (bold red line in graphs) 2020-06-05 to 2020-06-11

DateUKEUBEDEFRHUITNLPLPTROSE
2020-06-04 39904 127415 9548 8635 29010 539 33689 5990 1117 1455 1305 4562
2020-06-05 40100 128000 9570 8670 29100 540 33800 6000 1130 1460 1310 4590
2020-06-06 40300 128000 9590 8700 29100 550 33800 6010 1140 1470 1320 4630
2020-06-07 40500 128000 9610 8730 29200 550 33900 6020 1150 1480 1330 4660
2020-06-08 40700 129000 9630 8750 29200 550 34000 6030 1160 1490 1340 4690
2020-06-09 40900 129000 9650 8780 29300 560 34100 6040 1170 1500 1350 4720
2020-06-10 41100 129000 9670 8810 29400 560 34200 6050 1180 1510 1360 4760
2020-06-11 41400 130000 9690 8840 29400 570 34200 6060 1190 1520 1370 4790

Deaths count forecast (bold red line in graphs) 2020-06-05 to 2020-06-11

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NHUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-04 34021 7717 1356 1099 6363 1721 8071 12545 5031 984 5376 848 4630 108211 651 996 4444 1512 4007 385 2607 2118 583 5735 2052 458 2772 7190 2628 5521 1115 768 794 1006 273 11970 383 430 30174 2340 5832 397 1776 1445 1138 626
2020-06-05 35400 7800 1430 1140 6650 1740 8130 13400 5130 990 5520 900 4650 109000 660 1020 4510 1530 4030 390 2640 2140 590 5810 2070 470 2790 7230 2660 5550 1130 770 810 1020 280 12100 390 430 30300 2370 5890 400 1800 1460 1140 630
2020-06-06 36900 7900 1500 1170 6950 1770 8190 14300 5240 1000 5670 960 4670 110000 660 1030 4580 1540 4040 390 2670 2170 600 5890 2090 470 2810 7280 2690 5580 1150 780 820 1030 290 12200 400 440 30400 2410 5960 410 1820 1480 1150 640
2020-06-07 38500 7990 1580 1210 7260 1790 8250 15300 5360 1000 5830 1020 4700 111000 670 1050 4640 1550 4060 400 2710 2190 610 5970 2100 480 2820 7330 2730 5600 1170 780 830 1040 300 12300 400 440 30500 2450 6020 410 1840 1490 1150 650
2020-06-08 40100 8090 1650 1250 7580 1810 8310 16300 5480 1010 5990 1080 4720 112000 670 1080 4700 1570 4080 400 2740 2220 620 6050 2120 490 2840 7370 2760 5630 1190 790 850 1050 300 12500 410 440 30600 2480 6090 420 1870 1510 1160 660
2020-06-09 41700 8190 1740 1290 7920 1840 8370 17400 5610 1020 6160 1140 4740 113000 680 1100 4770 1580 4100 410 2770 2240 630 6130 2140 490 2860 7420 2800 5660 1210 790 860 1060 310 12600 410 450 30700 2520 6150 430 1890 1520 1160 660
2020-06-10 43500 8290 1820 1340 8280 1860 8430 18600 5730 1020 6330 1200 4760 114000 680 1120 4830 1590 4120 410 2810 2270 640 6210 2150 500 2880 7470 2830 5680 1230 790 870 1070 320 12800 420 450 30800 2560 6220 430 1910 1540 1170 670
2020-06-11 45300 8390 1910 1380 8650 1890 8490 19900 5860 1030 6510 1270 4780 114000 690 1140 4900 1610 4130 420 2840 2290 650 6300 2170 510 2890 7520 2870 5710 1250 800 890 1080 330 13000 430 450 30900 2600 6290 440 1940 1550 1170 680

Deaths count scenario forecast (bold purple line in graphs) 2020-06-05 to 2020-06-13

DateUKEUBEDEFRHUITNLPLPTROSEBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NHUS-NJUS-NMUS-NVUS-NYUS-OHUS-PAUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-04 39904 127415 9548 8635 29010 539 33689 5990 1117 1455 1305 4562 34021 7717 1356 1099 6363 1721 8071 12545 5031 984 5376 848 4630 108211 651 996 4444 1512 4007 385 2607 2118 583 5735 2052 458 2772 7190 2628 5521 1115 768 794 1006 273 11970 383 430 30174 2340 5832 397 1776 1445 1138 626
2020-06-05 40200 128000 9560 8650 29100 540 33800 6000 1130 1470 1310 4590 35000 7770 1430 1150 6600 1740 8130 13100 5120 990 5520 890 4650 109000 660 1010 4500 1520 4030 390 2630 2140 590 5750 2070 460 2780 7260 2650 5550 1130 780 810 1020 280 12000 390 430 30200 2370 5870 400 1790 1460 1140 630
2020-06-06 40500 128000 9580 8670 29100 540 33800 6010 1130 1470 1320 4600 36100 7840 1510 1190 6830 1760 8190 13900 5240 1000 5690 940 4670 110000 660 1030 4560 1530 4040 390 2660 2150 600 5820 2080 470 2790 7310 2680 5580 1140 780 820 1040 290 12100 390 440 30300 2410 5920 400 1810 1470 1140 640
2020-06-07 40700 128000 9600 8680 29100 540 33900 6020 1140 1480 1320 4630 37400 7900 1590 1240 7050 1780 8240 14900 5350 1000 5840 990 4680 110000 670 1050 4600 1540 4060 390 2680 2170 610 5870 2100 470 2810 7360 2700 5600 1160 780 830 1060 290 12100 400 440 30300 2440 5980 400 1820 1480 1150 650
2020-06-08 40900 128000 9610 8700 29200 550 33900 6030 1150 1490 1320 4640 38700 7950 1670 1280 7280 1800 8300 16100 5470 1010 5980 1040 4700 111000 670 1060 4640 1540 4070 390 2700 2190 620 5920 2110 480 2810 7400 2720 5620 1180 790 840 1080 300 12200 400 440 30400 2470 6020 400 1840 1500 1150 660
2020-06-09 41000 128000 9630 8710 29200 550 34000 6040 1150 1490 1330 4660 39900 8000 1750 1320 7490 1820 8340 17000 5580 1020 6140 1090 4710 111000 680 1070 4670 1550 4090 400 2710 2200 620 5960 2120 490 2820 7420 2740 5630 1190 790 850 1100 320 12300 400 440 30500 2490 6050 400 1850 1510 1150 670
2020-06-10 41200 129000 9640 8730 29300 550 34100 6050 1160 1500 1330 4680 41200 8040 1820 1330 7690 1840 8390 17800 5670 1020 6280 1140 4720 112000 680 1080 4700 1550 4100 400 2730 2210 630 6000 2140 490 2830 7450 2760 5650 1210 790 850 1110 320 12300 410 450 30500 2510 6080 400 1860 1520 1150 680
2020-06-11 41300 129000 9650 8740 29300 550 34100 6050 1160 1500 1330 4700 42300 8100 1900 1350 7890 1850 8430 18700 5750 1030 6420 1190 4730 112000 690 1080 4730 1560 4110 400 2750 2220 640 6030 2150 490 2830 7470 2780 5670 1220 800 860 1130 330 12300 410 450 30600 2530 6110 400 1870 1540 1150 690
2020-06-12 41500 129000 9660 8760 29300 550 34200 6060 1160 1510 1340 4730 43600 8140 1970 1390 8070 1870 8480 19700 5860 1040 6580 1240 4740 113000 690 1090 4750 1560 4120 400 2760 2230 640 6070 2160 500 2840 7480 2800 5680 1240 800 870 1140 340 12400 410 450 30600 2550 6130 410 1880 1550 1150 700
2020-06-13 41600 129000 9680 8780 29400 550 34200 6060 1160 1510 1340 4740 44800 8180 2030 1410 8250 1890 8520 21200 5950 1040 6690 1300 4750 113000 690 1100 4770 1570 4140 410 2780 2240 650 6110 2160 500 2850 7510 2810 5690 1250 810 880 1150 340 12400 420 450 30700 2560 6160 410 1890 1560 1150 700

Peak increase in estimated trend of Deaths in Europe 2020-06-04

UKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
Peak date04-0904-0704-0804-1004-1104-0804-1504-0404-0204-2104-0704-1804-2403-2804-0704-2404-1105-0104-2204-0704-04
Peak daily increment 1011 3234 23 354 7 11 270 18 899 18 1017 14 95 832 160 29 33 25 118 9 63
Days from 100 to peak 22 34 9 17 1 -1 23 3 20 1 23 5 22 24 18 18 13 29 25 -2 12
Days from peak/2 to peak 16 24 17 15 22 13 22 14 16 23 16 25 25 19 18 28 19 37 27 20 16
Last total 39904 127415 670 9548 250 326 8635 582 27133 322 29010 539 1664 33689 5990 1117 1455 1305 4562 238 1921
Last daily increment 176 264 0 26 0 1 33 2 5 1 43 5 5 88 13 2 8 9 20 1 0
Last week 1743 1629 2 118 5 7 131 14 12 8 347 22 19 460 59 66 72 57 212 2 2
Days since peak 56 58 57 55 54 57 50 61 63 44 58 47 41 68 58 41 54 34 43 58 61

Peak increase in estimated trend of Deaths 2020-06-04

BrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KSUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NHUS-NJUS-NMUS-NVUS-NYUS-OHUS-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
Peak date --05-05 -- -- --04-1403-19 --05-2704-1205-29 --04-1904-1505-0605-0804-2304-2404-2004-2505-1405-0704-2005-2005-1304-2904-1404-1604-1404-2504-2904-2205-2805-0805-0605-2805-2904-3005-1404-1204-0704-2904-2105-0504-2904-0305-1404-2204-0605-29
Peak daily increment 178 38 148 150 29 185 125 2270 17 38 88 52 116 12 11 54 45 16 128 50 7 12 76 245 66 146 27 20 23 25 14 315 11 11 1072 56 9 170 20 10 40 53 31 14
Days from 100 to peak 34 16 14 50 11 48 22 30 23 26 26 21 18 5 20 36 20 26 43 25 -6 1 18 24 22 25 41 27 23 45 16 29 17 0 18 25 8 32 14 -8 40 7 14 51
Days from peak/2 to peak 35 26 16 52 25 48 24 21 37 36 30 28 20 26 45 42 25 48 44 32 18 20 21 23 26 26 55 41 38 57 31 29 42 17 13 31 27 33 33 5 48 13 30 65
Last total 34021 7717 1356 1099 6363 1721 8071 12545 5031 984 5376 848 4630 108211 651 996 4444 1512 4007 475 385 2607 2118 583 5735 2052 227 458 2772 7190 2628 5521 1115 768 794 1006 273 11970 383 430 30174 2340 344 5832 501 397 1776 1445 1138 626
Last daily increment 1473 138 81 42 275 23 59 816 137 10 168 56 21 1036 0 16 70 18 18 2 10 41 26 9 115 20 1 8 13 51 40 25 29 4 12 7 8 90 8 1 155 40 3 90 0 13 32 17 3 10
Last week 6143 654 412 244 1383 201 394 3130 801 42 1002 237 141 5402 41 114 367 76 139 15 29 194 162 59 466 106 12 40 110 479 217 185 119 64 84 87 35 439 39 24 528 209 15 368 18 40 157 87 27 58
Days since peak 30 51 77 8 53 6 46 50 29 27 42 41 45 40 21 28 45 15 22 36 51 49 51 40 36 43 7 27 29 7 6 35 21 53 58 36 44 30 36 62 21 43 59 6

Initial visual evaluation of forecasts of Deaths