COVID-19 short-term forecasts Deaths 2020-06-29


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 The documentation that is provided is still in progress and not peer reviewed.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.
[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.
[2020-06-06] Removed Brazil from yesterday's forecasts (only; last observation 2020-06-05).
[2020-06-24] Research presentation on short-term COVID-19 forecasting on 26 June (14:00 UK time) at the Quarterly Forecasting Forum of the IIF UK Chapter.
[2020-06-29] Tables in April included the world, but not the world as we know it (double counting China and the US). So removed the world from those old tables.
Why short-term forecasts can be better than models for predicting how pandemics evolve just appeared at The Conversation.
Thursday 2 July webinar at the FGV EESP - São Paolo School of Economics. This starts at 16:00 UK time (UTC+01:00) and will be streamed here.

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) 2020-06-30 to 2020-07-06

DateUKEUDEESFRITPLPTROSE
2020-06-29 43575 132876 8976 28346 29736 34744 1444 1568 1634 5310
2020-06-30 43700 133000 8980 28400 29700 34800 1450 1570 1640 5320
2020-07-01 43800 133000 9000 28400 29800 34800 1460 1580 1650 5340
2020-07-02 43900 133000 9010 28400 29800 34800 1470 1580 1660 5360
2020-07-03 44000 133000 9020 28400 29800 34800 1480 1590 1680 5390
2020-07-04 44100 133000 9030 28400 29800 34800 1490 1590 1690 5410
2020-07-05 44300 134000 9040 28400 29900 34900 1500 1590 1710 5430
2020-07-06 44400 134000 9060 28500 29900 34900 1510 1600 1720 5450

Deaths count average forecast (bold black line in graphs) 2020-06-30 to 2020-07-06

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-CTUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NHUS-NJUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-29 58314 8628 5575 3256 16893 2805 10670 27121 9504 1255 9152 2529 5115 126140 905 265 1597 5983 4320 3447 2737 711 6902 2432 560 3091 8087 3151 6088 1435 973 1059 1357 367 14992 31403 2818 6614 720 585 2416 1740 1320 777
2020-06-30 59000 8640 5670 3370 17100 2830 10700 27500 9700 1260 9300 2580 5130 127000 910 270 1620 6020 4330 3480 2750 710 6920 2440 560 3100 8110 3160 6100 1440 980 1070 1360 370 15100 31400 2830 6630 720 590 2440 1750 1320 780
2020-07-01 59900 8650 5820 3510 17400 2870 10800 28100 9900 1270 9400 2640 5150 127000 920 270 1650 6070 4330 3510 2760 720 6950 2450 570 3110 8140 3180 6110 1440 980 1080 1370 370 15200 31500 2840 6650 730 600 2460 1770 1330 780
2020-07-02 60700 8660 5980 3680 17700 2910 11000 28600 10000 1280 9500 2700 5170 128000 920 280 1670 6120 4340 3540 2780 720 6980 2460 570 3120 8170 3190 6110 1450 990 1090 1380 380 15400 31500 2860 6670 740 600 2490 1780 1340 790
2020-07-03 61600 8680 6130 3850 18100 2950 11100 29200 10200 1290 9600 2770 5190 128000 930 280 1700 6170 4350 3570 2790 720 7010 2470 570 3140 8200 3200 6120 1460 990 1100 1390 380 15500 31500 2870 6700 740 610 2510 1790 1340 790
2020-07-04 62400 8690 6300 4020 18400 3000 11200 29800 10400 1300 9700 2830 5210 129000 940 290 1740 6220 4360 3600 2810 730 7040 2480 580 3150 8240 3220 6130 1460 990 1110 1400 390 15700 31600 2880 6720 750 620 2540 1810 1350 800
2020-07-05 63300 8710 6460 4210 18800 3040 11300 30400 10600 1310 9900 2900 5230 130000 940 290 1770 6270 4370 3640 2820 730 7070 2490 580 3160 8270 3230 6140 1470 1000 1120 1410 390 15900 31600 2900 6740 760 620 2560 1820 1350 800
2020-07-06 64200 8730 6640 4410 19200 3080 11400 31000 10800 1320 10000 2970 5250 130000 950 300 1800 6320 4370 3670 2840 730 7100 2500 590 3170 8300 3250 6150 1470 1000 1130 1420 400 16000 31700 2910 6770 760 630 2590 1840 1360 800

Deaths count forecast Europe (bold red line in graphs) 2020-06-30 to 2020-07-06

DateUKEUDEESFRITPLPTROSE
2020-06-29 43575 132876 8976 28346 29736 34744 1444 1568 1634 5310
2020-06-30 43700 133000 8990 28400 29800 34800 1450 1570 1650 5330
2020-07-01 43800 133000 9000 28500 29800 34800 1460 1580 1660 5350
2020-07-02 43900 133000 9010 28500 29800 34800 1470 1580 1680 5370
2020-07-03 44000 133000 9020 28600 29800 34800 1480 1590 1700 5390
2020-07-04 44100 133000 9030 28700 29800 34900 1480 1590 1710 5400
2020-07-05 44200 134000 9040 28700 29900 34900 1490 1600 1730 5420
2020-07-06 44300 134000 9050 28800 29900 34900 1500 1600 1750 5440

Deaths count forecast (bold red line in graphs) 2020-06-30 to 2020-07-06

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-CTUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NHUS-NJUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-29 58314 8628 5575 3256 16893 2805 10670 27121 9504 1255 9152 2529 5115 126140 905 265 1597 5983 4320 3447 2737 711 6902 2432 560 3091 8087 3151 6088 1435 973 1059 1357 367 14992 31403 2818 6614 720 585 2416 1740 1320 777
2020-06-30 58900 8660 5710 3430 17300 2840 10800 27500 9700 1270 9260 2580 5130 127000 910 270 1610 6030 4330 3470 2750 710 6930 2440 560 3100 8120 3160 6100 1440 980 1070 1370 370 15200 31400 2830 6630 730 590 2430 1750 1330 780
2020-07-01 59600 8690 5850 3570 17700 2880 10900 28000 9900 1280 9380 2630 5150 127000 920 270 1620 6070 4330 3500 2760 720 6950 2450 570 3110 8150 3180 6110 1450 980 1080 1370 370 15300 31500 2840 6650 730 600 2450 1760 1330 780
2020-07-02 60200 8710 5980 3720 18100 2920 11100 28400 10000 1290 9490 2680 5170 128000 920 280 1630 6110 4340 3530 2780 720 6970 2460 570 3120 8180 3190 6120 1460 980 1100 1380 370 15500 31500 2850 6670 740 600 2470 1770 1340 790
2020-07-03 60800 8740 6120 3880 18400 2960 11200 28900 10200 1300 9600 2740 5190 129000 930 280 1650 6160 4350 3550 2790 730 7000 2460 570 3130 8210 3200 6120 1460 990 1110 1390 380 15700 31500 2870 6690 740 610 2480 1780 1340 790
2020-07-04 61400 8770 6260 4040 18900 3000 11300 29300 10400 1310 9720 2790 5200 129000 940 290 1660 6200 4350 3580 2800 730 7020 2470 570 3140 8240 3210 6130 1470 990 1120 1390 380 15900 31600 2880 6710 750 610 2500 1790 1350 790
2020-07-05 62000 8790 6400 4210 19300 3040 11400 29800 10600 1320 9840 2850 5220 130000 940 290 1670 6240 4360 3610 2810 730 7040 2480 580 3150 8270 3220 6140 1480 990 1130 1400 380 16000 31600 2890 6730 750 620 2520 1800 1350 790
2020-07-06 62700 8820 6550 4390 19700 3080 11600 30200 10800 1330 9950 2910 5240 130000 950 300 1680 6290 4370 3640 2820 740 7070 2490 580 3160 8300 3230 6150 1480 1000 1150 1400 380 16200 31600 2900 6750 760 630 2530 1810 1360 800

Deaths count scenario forecast (bold purple line in graphs) 2020-06-30 to 2020-07-08

DateUKEUDEESFRITPLPTROSEBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-CTUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NHUS-NJUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-29 43575 132876 8976 28346 29736 34744 1444 1568 1634 5310 58314 8628 5575 3256 16893 2805 10670 27121 9504 1255 9152 2529 5115 126140 905 265 1597 5983 4320 3447 2737 711 6902 2432 560 3091 8087 3151 6088 1435 973 1059 1357 367 14992 31403 2818 6614 720 585 2416 1740 1320 777
2020-06-30 43800 133000 8980 28300 29800 34800 1450 1570 1630 5330 59000 8630 5650 3450 17100 2840 10700 27600 9600 1270 9250 2600 5130 127000 910 270 1630 6020 4330 3470 2760 710 6930 2440 560 3100 8120 3160 6100 1440 980 1070 1370 370 15100 31400 2830 6640 730 590 2440 1760 1320 780
2020-07-01 43900 133000 8990 28300 29800 34800 1460 1580 1650 5330 59700 8640 5740 3600 17300 2880 10800 28100 9800 1270 9360 2670 5150 128000 920 280 1670 6040 4330 3490 2770 710 6950 2450 570 3110 8140 3170 6100 1440 980 1080 1370 370 15600 31500 2840 6660 730 600 2460 1770 1330 790
2020-07-02 44000 133000 9000 28300 29800 34800 1460 1580 1660 5340 60400 8640 5850 3720 17600 2930 10900 28800 10000 1280 9460 2730 5160 128000 930 290 1710 6060 4340 3500 2780 720 6970 2460 570 3120 8170 3180 6110 1450 980 1090 1380 380 16000 31500 2850 6680 730 600 2460 1780 1330 790
2020-07-03 44100 133000 9010 28300 29800 34800 1460 1580 1670 5350 60900 8640 5970 3830 17800 2970 11000 29400 10100 1290 9550 2800 5180 129000 930 290 1750 6090 4350 3520 2790 720 6990 2460 570 3130 8190 3190 6110 1450 980 1100 1390 380 16100 31500 2850 6690 730 610 2480 1800 1340 790
2020-07-04 44100 133000 9010 28300 29800 34800 1470 1590 1680 5360 61500 8660 6060 4110 18000 3010 11100 29900 10200 1300 9630 2860 5190 129000 940 290 1790 6110 4350 3530 2800 720 7000 2470 580 3140 8210 3200 6120 1450 990 1110 1390 380 16300 31500 2860 6700 740 610 2490 1810 1340 790
2020-07-05 44200 133000 9020 28300 29800 34800 1470 1590 1680 5370 62000 8680 6120 4250 18200 3060 11200 30100 10300 1310 9700 2950 5200 130000 940 290 1840 6120 4360 3540 2810 720 7020 2480 580 3140 8230 3210 6120 1460 990 1120 1400 380 16600 31600 2870 6710 740 620 2500 1820 1340 800
2020-07-06 44300 133000 9020 28300 29900 34800 1470 1590 1690 5370 62500 8700 6200 4380 18300 3110 11300 30500 10400 1310 9770 2990 5210 130000 940 290 1880 6140 4360 3560 2820 720 7040 2480 580 3150 8240 3220 6130 1460 990 1130 1410 380 16900 31600 2880 6720 740 620 2510 1830 1340 800
2020-07-07 44400 133000 9030 28300 29900 34900 1480 1590 1690 5380 63000 8710 6260 4530 18500 3150 11300 30900 10500 1320 9830 3010 5220 130000 950 290 1920 6160 4370 3570 2830 730 7050 2490 580 3150 8260 3230 6130 1460 990 1140 1410 390 17300 31600 2880 6730 740 630 2520 1840 1350 800
2020-07-08 44400 134000 9040 28300 29900 34900 1490 1590 1700 5380 63500 8720 6320 4680 18800 3180 11400 31300 10700 1330 9890 3050 5230 131000 950 290 1960 6180 4370 3580 2840 730 7070 2500 590 3160 8270 3240 6140 1470 1000 1150 1420 390 17600 31600 2890 6740 750 630 2530 1850 1350 800

Peak increase in estimated trend of Deaths in Europe 2020-06-29

UKEUATBEBGBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
Peak date04-0904-0704-0704-1006-0504-1104-0904-1504-0403-3004-2104-0704-1904-2403-2804-0704-2404-1105-0104-2204-0804-04
Peak daily increment 1019 3232 22 349 4 8 11 269 18 896 17 1028 14 101 839 159 29 33 25 120 9 61
Days from 100 to peak 22 34 8 17 21 1 0 23 3 17 1 23 6 22 24 18 18 13 29 25 -1 12
Days from peak/2 to peak 16 24 16 15 80 22 14 22 14 13 24 16 26 25 19 18 28 19 37 27 21 17
Last total 43575 132876 703 9732 223 257 348 8976 605 28346 328 29736 585 1735 34744 6107 1444 1568 1634 5310 249 1962
Last daily increment 25 123 1 0 4 0 0 8 1 3 0 32 4 0 6 2 6 4 22 30 0 0
Last week 648 676 10 19 15 1 9 62 2 21 1 84 12 15 69 12 69 28 95 149 1 6
Days since peak 81 83 83 80 24 79 81 75 86 91 69 83 71 66 93 83 66 79 59 68 82 86

Peak increase in estimated trend of Deaths 2020-06-29

BrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-ILUS-INUS-KSUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NEUS-NHUS-NJUS-NMUS-NVUS-NYUS-OHUS-OKUS-ORUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
Peak date06-0405-0406-08 --06-1606-1503-1906-2406-1404-1205-2906-2304-2004-1505-0606-1806-2404-2304-2404-2004-2506-2305-0604-2005-2005-1304-2904-1404-1604-1404-2704-2904-2205-2805-0505-0605-2905-2005-2906-2505-1204-1204-0704-2904-2104-0905-0504-2904-0605-1504-2204-0605-29
Peak daily increment 1097 179 309 930 49 147 836 245 27 184 82 125 2272 17 7 38 87 51 115 12 25 54 46 16 127 50 7 11 75 221 63 151 27 19 22 27 23 15 987 10 10 1068 54 9 4 170 18 10 39 47 31 14
Days from 100 to peak 68 33 53 72 78 14 79 68 11 48 55 23 30 23 31 73 26 21 18 5 60 35 20 26 43 25 -6 1 18 26 22 25 41 26 23 46 0 16 86 15 0 18 25 8 -21 32 14 -5 41 7 14 51
Days from peak/2 to peak 59 34 47 60 87 16 70 67 25 48 68 25 21 37 79 84 30 28 20 26 79 41 25 48 44 32 18 21 21 26 26 26 55 39 38 58 41 31 83 40 18 13 31 27 21 33 34 8 50 14 30 65
Last total 58314 8628 5575 3256 16893 2805 10670 27121 9504 1255 9152 2529 5115 126140 905 265 1597 5983 1681 4320 551 507 3447 2737 711 6902 2432 272 560 3091 8087 3151 6088 1435 973 1059 1357 268 367 14992 493 504 31403 2818 385 204 6614 720 585 2416 1740 1320 777
Last daily increment 692 46 66 0 418 51 162 473 187 11 92 73 18 337 7 1 4 51 5 4 1 0 28 4 5 14 5 4 2 5 35 9 4 10 0 20 5 2 0 17 1 4 6 11 0 2 8 4 8 14 8 10 0
Last week 5669 116 1070 732 2417 270 807 3744 1100 69 803 427 114 4909 51 28 202 357 16 43 14 3 209 85 22 195 55 13 23 70 206 84 52 42 28 70 61 13 24 1967 17 12 171 83 14 12 150 47 51 180 95 36 27
Days since peak 25 56 21 13 14 102 5 15 78 31 6 70 75 54 11 5 67 66 70 65 6 54 70 40 47 61 76 74 76 63 61 68 32 55 54 31 40 31 4 48 78 83 61 69 81 55 61 84 45 68 84 31

Initial visual evaluation of forecasts of Deaths