COVID-19 short-term forecasts Deaths 2020-06-15


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 The documentation that is provided is still in progress and not peer reviewed.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.
[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 (last observation 2020-06-05).

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) 2020-06-16 to 2020-06-22

DateUKEUBEDEFRITNLPLPTROSE
2020-06-15 41736 129624 9661 8807 29375 34371 6065 1256 1520 1427 4891
2020-06-16 41800 130000 9670 8820 29400 34400 6070 1270 1520 1430 4910
2020-06-17 42000 130000 9680 8830 29400 34500 6080 1280 1530 1440 4930
2020-06-18 42100 130000 9690 8850 29500 34500 6090 1290 1530 1450 4960
2020-06-19 42300 130000 9700 8860 29500 34600 6090 1300 1540 1470 4980
2020-06-20 42400 131000 9710 8880 29500 34700 6100 1310 1540 1480 5010
2020-06-21 42600 131000 9710 8890 29600 34700 6110 1320 1550 1490 5030
2020-06-22 42800 131000 9720 8910 29600 34800 6110 1330 1550 1510 5050

Deaths count average forecast (bold black line in graphs) 2020-06-16 to 2020-06-22

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NEUS-NHUS-NJUS-NMUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-15 43959 8228 3362 1808 9900 2198 8950 17580 6860 1098 7081 1568 4825 116127 769 1201 5114 1605 4204 515 2938 2452 658 6325 2251 505 2906 7636 2919 5945 1304 858 895 1140 219 320 12708 440 30856 2575 6243 602 475 2001 1552 1220 694
2020-06-16 44600 8260 3520 1830 10300 2250 8980 17900 6990 1110 7210 1630 4840 117000 770 1220 5150 1610 4220 520 2960 2470 660 6360 2260 510 2910 7670 2940 5950 1320 860 900 1150 220 320 12800 440 30900 2590 6270 610 480 2020 1560 1230 700
2020-06-17 45500 8300 3710 1890 10700 2310 9050 18400 7190 1120 7360 1710 4850 117000 780 1230 5210 1620 4240 520 2980 2500 670 6400 2280 510 2920 7710 2960 5970 1330 870 910 1160 220 330 12900 450 31000 2610 6310 610 490 2040 1570 1230 700
2020-06-18 46300 8340 3900 1960 11200 2370 9120 18800 7420 1130 7510 1790 4870 118000 790 1250 5260 1630 4250 520 3010 2530 670 6450 2290 510 2930 7750 2980 5980 1340 870 920 1170 230 330 13000 450 31100 2630 6340 620 500 2050 1580 1240 710
2020-06-19 47100 8390 4110 2030 11700 2440 9200 19200 7670 1140 7660 1870 4890 119000 800 1270 5320 1630 4270 530 3040 2560 680 6500 2310 520 2940 7780 3010 5990 1360 880 920 1180 230 340 13100 460 31100 2660 6380 620 500 2070 1590 1250 710
2020-06-20 47900 8430 4330 2110 12200 2500 9280 19600 7920 1150 7820 1960 4900 119000 800 1290 5380 1640 4290 530 3070 2590 680 6550 2320 520 2950 7820 3030 6000 1370 880 930 1190 230 340 13300 460 31200 2680 6410 630 510 2090 1600 1260 720
2020-06-21 48800 8470 4560 2190 12700 2570 9350 20100 8180 1160 7980 2050 4920 120000 810 1310 5440 1650 4300 530 3090 2620 690 6600 2340 520 2970 7860 3050 6010 1390 880 940 1200 240 350 13400 470 31300 2700 6450 630 520 2110 1600 1260 720
2020-06-22 49700 8520 4820 2270 13300 2640 9430 20600 8450 1170 8140 2140 4930 121000 820 1330 5490 1660 4320 540 3120 2650 690 6650 2350 530 2980 7900 3080 6030 1400 890 950 1210 240 350 13600 480 31400 2720 6480 640 530 2130 1610 1270 730

Deaths count forecast Europe (bold red line in graphs) 2020-06-16 to 2020-06-22

DateUKEUBEDEFRITNLPLPTROSE
2020-06-15 41736 129624 9661 8807 29375 34371 6065 1256 1520 1427 4891
2020-06-16 41900 130000 9670 8820 29400 34400 6070 1270 1520 1440 4920
2020-06-17 42000 130000 9680 8830 29400 34500 6080 1270 1530 1450 4940
2020-06-18 42100 130000 9690 8850 29500 34500 6090 1280 1530 1470 4960
2020-06-19 42200 130000 9690 8860 29500 34600 6090 1290 1540 1490 4980
2020-06-20 42300 130000 9700 8870 29500 34600 6100 1300 1540 1500 5000
2020-06-21 42500 131000 9710 8880 29600 34700 6110 1310 1540 1520 5020
2020-06-22 42600 131000 9720 8890 29600 34800 6110 1320 1550 1540 5040

Deaths count forecast (bold red line in graphs) 2020-06-16 to 2020-06-22

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NEUS-NHUS-NJUS-NMUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-15 43959 8228 3362 1808 9900 2198 8950 17580 6860 1098 7081 1568 4825 116127 769 1201 5114 1605 4204 515 2938 2452 658 6325 2251 505 2906 7636 2919 5945 1304 858 895 1140 219 320 12708 440 30856 2575 6243 602 475 2001 1552 1220 694
2020-06-16 44600 8260 3510 1870 10300 2260 9030 18000 7110 1110 7220 1660 4840 117000 780 1210 5150 1610 4220 520 2960 2480 660 6360 2270 510 2910 7670 2940 5960 1320 860 900 1150 220 320 12800 450 30900 2590 6270 610 480 2020 1560 1230 700
2020-06-17 45200 8290 3650 1950 10800 2320 9110 18400 7360 1120 7360 1750 4860 117000 780 1220 5190 1620 4240 520 2980 2510 670 6400 2280 510 2920 7710 2960 5960 1330 870 910 1160 230 320 12900 450 31000 2610 6310 610 490 2040 1570 1230 700
2020-06-18 45700 8320 3800 2030 11200 2380 9200 18800 7630 1140 7500 1840 4870 118000 780 1230 5220 1630 4250 520 2990 2530 670 6430 2290 520 2930 7740 2980 5970 1340 870 910 1170 230 330 13000 460 31100 2630 6340 620 500 2050 1580 1240 710
2020-06-19 46300 8350 3950 2120 11700 2450 9280 19200 7910 1150 7640 1940 4890 118000 790 1250 5260 1630 4260 530 3010 2560 680 6470 2300 520 2940 7780 2990 5980 1350 870 920 1170 230 330 13100 460 31100 2640 6370 620 500 2070 1580 1250 710
2020-06-20 46900 8380 4110 2210 12200 2520 9360 19600 8200 1160 7790 2040 4900 119000 790 1260 5290 1640 4270 530 3030 2590 680 6510 2320 520 2940 7810 3010 5990 1360 880 930 1180 240 330 13300 470 31200 2660 6400 620 510 2090 1590 1250 710
2020-06-21 47500 8410 4280 2310 12700 2590 9440 20000 8510 1170 7930 2150 4920 120000 800 1270 5330 1650 4290 530 3050 2620 680 6540 2330 530 2950 7850 3030 5990 1370 880 930 1190 240 330 13400 470 31200 2680 6430 630 520 2100 1600 1260 720
2020-06-22 48100 8450 4450 2410 13300 2660 9530 20400 8820 1180 8080 2260 4930 120000 800 1280 5360 1650 4300 530 3070 2640 690 6580 2340 530 2960 7880 3050 6000 1380 880 940 1200 250 330 13600 480 31300 2700 6460 630 530 2120 1610 1270 720

Deaths count scenario forecast (bold purple line in graphs) 2020-06-16 to 2020-06-24

DateUKEUBEDEFRITNLPLPTROSEBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-AZUS-CAUS-COUS-CTUS-DCUS-FLUS-GAUS-IAUS-ILUS-INUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NEUS-NHUS-NJUS-NMUS-NYUS-OHUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
2020-06-15 41736 129624 9661 8807 29375 34371 6065 1256 1520 1427 4891 43959 8228 3362 1808 9900 2198 8950 17580 6860 1098 7081 1568 4825 116127 769 1201 5114 1605 4204 515 2938 2452 658 6325 2251 505 2906 7636 2919 5945 1304 858 895 1140 219 320 12708 440 30856 2575 6243 602 475 2001 1552 1220 694
2020-06-16 41900 130000 9670 8820 29400 34500 6070 1260 1530 1440 4920 44700 8280 3580 1840 10300 2230 9000 17900 7030 1110 7200 1640 4840 117000 770 1210 5170 1610 4220 520 2960 2470 660 6370 2260 510 2920 7680 2940 5960 1320 860 910 1150 220 330 12800 450 30900 2600 6280 610 480 2020 1560 1230 700
2020-06-17 42100 130000 9680 8830 29400 34500 6070 1270 1530 1450 4950 45500 8310 3770 1900 10600 2270 9060 18200 7220 1120 7320 1710 4850 117000 780 1220 5200 1620 4240 520 2970 2490 670 6410 2270 510 2930 7710 2960 5970 1330 870 910 1160 220 330 12800 450 31000 2620 6310 610 490 2030 1570 1240 700
2020-06-18 42200 130000 9680 8840 29500 34500 6080 1280 1540 1460 4960 46000 8340 3920 1950 11000 2320 9110 18500 7450 1120 7440 1770 4870 118000 780 1230 5230 1630 4260 530 2980 2510 670 6450 2280 510 2930 7740 2980 5980 1340 870 920 1160 230 330 12900 450 31000 2630 6350 610 500 2050 1580 1240 710
2020-06-19 42300 130000 9690 8850 29500 34600 6080 1290 1540 1470 4980 46700 8380 4090 2000 11300 2360 9150 18800 7620 1130 7570 1840 4880 118000 780 1240 5260 1630 4270 530 2990 2530 680 6480 2290 520 2940 7770 2990 6000 1350 870 920 1170 230 330 12900 460 31100 2660 6380 620 510 2060 1590 1250 710
2020-06-20 42400 130000 9700 8860 29500 34600 6090 1290 1550 1480 5000 47300 8410 4270 2060 11700 2420 9190 19100 7820 1130 7670 1890 4890 119000 780 1240 5290 1640 4280 530 3000 2540 680 6510 2300 520 2950 7790 3010 6010 1360 880 930 1180 240 340 13000 460 31100 2670 6410 620 510 2070 1590 1250 710
2020-06-21 42500 130000 9700 8870 29500 34600 6090 1300 1550 1480 5020 47800 8440 4430 2120 12000 2470 9220 19300 8000 1140 7780 1950 4900 119000 780 1250 5300 1650 4300 530 3010 2560 690 6530 2310 520 2960 7820 3020 6020 1360 880 930 1180 240 340 13000 460 31200 2690 6430 620 520 2080 1600 1260 720
2020-06-22 42600 131000 9710 8880 29500 34700 6100 1300 1550 1490 5040 48300 8460 4580 2160 12300 2530 9260 19600 8200 1140 7870 2010 4910 119000 790 1260 5320 1650 4310 530 3020 2570 690 6560 2320 520 2970 7840 3040 6030 1370 880 940 1180 240 340 13000 460 31200 2710 6450 620 530 2090 1610 1260 720
2020-06-23 42700 131000 9720 8890 29600 34700 6100 1310 1560 1500 5060 48800 8480 4740 2210 12700 2580 9280 19800 8370 1140 7960 2060 4920 120000 790 1260 5340 1660 4320 530 3030 2580 690 6590 2320 530 2970 7860 3050 6040 1380 890 940 1190 250 340 13100 460 31300 2720 6460 630 540 2100 1620 1270 720
2020-06-24 42900 131000 9730 8900 29600 34700 6100 1310 1560 1500 5070 49300 8500 4870 2250 13000 2640 9300 20000 8540 1140 8030 2110 4930 120000 790 1270 5360 1660 4330 540 3050 2590 700 6610 2330 530 2980 7880 3060 6050 1390 890 950 1200 250 350 13100 470 31300 2740 6480 630 550 2110 1620 1270 730

Peak increase in estimated trend of Deaths in Europe 2020-06-15

UKEUATBEBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
Peak date04-0904-0704-0704-1104-1104-0904-1504-0503-3004-2104-0704-1904-2403-2804-0704-2504-1105-0104-2204-0804-04
Peak daily increment 1025 3257 23 346 8 11 274 18 899 18 1037 14 94 843 162 29 33 25 117 9 61
Days from 100 to peak 22 34 8 18 1 0 23 4 17 1 23 6 22 24 18 19 13 29 25 -1 12
Days from peak/2 to peak 16 24 16 16 22 14 22 15 13 23 16 26 25 19 18 29 19 37 27 21 17
Last total 41736 129624 678 9661 255 330 8807 598 27136 326 29375 563 1706 34371 6065 1256 1520 1427 4891 242 1939
Last daily increment 38 127 1 6 1 1 6 1 0 0 29 1 0 26 6 9 3 17 17 0 1
Last week 853 1021 6 42 4 2 71 5 0 2 138 13 15 328 34 73 28 73 174 3 5
Days since peak 67 69 69 65 65 67 61 71 77 55 69 57 52 79 69 51 65 45 54 68 72

Peak increase in estimated trend of Deaths 2020-06-15

BrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-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-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
Peak date06-0305-0406-08 -- -- --03-1906-03 --04-1205-29 --04-2004-1505-0605-0804-2304-2404-2004-2505-1605-0604-2005-2005-1304-2904-1404-2104-1404-2504-2904-2205-2805-0505-0605-2905-2005-2904-1505-1404-1204-0704-2904-2105-0604-2904-0305-1504-2304-0605-29
Peak daily increment 1094 178 295 148 614 27 185 124 2287 17 37 87 50 115 12 11 53 44 16 129 50 7 12 73 228 64 149 26 19 22 27 20 16 303 10 10 1066 56 10 171 19 10 40 48 30 14
Days from 100 to peak 67 33 53 14 58 11 48 23 30 23 26 26 21 18 5 22 35 20 26 43 25 -6 6 18 24 22 25 41 26 23 46 0 16 14 17 0 18 25 8 33 14 -8 41 8 14 51
Days from peak/2 to peak 58 34 48 16 51 25 48 25 21 37 37 30 28 20 26 47 41 25 48 44 32 18 25 21 23 26 26 55 39 38 58 42 30 14 42 18 13 31 26 34 33 5 49 15 30 65
Last total 43959 8228 3362 1808 9900 2198 8950 17580 6860 1098 7081 1568 4825 116127 769 1201 5114 1605 4204 515 423 2938 2452 658 6325 2251 246 505 2906 7636 2919 5945 1304 858 895 1140 219 320 12708 440 465 30856 2575 359 6243 602 475 2001 1552 1220 694
Last daily increment 627 10 39 138 380 64 113 439 172 10 143 88 18 395 1 12 15 7 3 0 1 7 39 6 18 11 1 6 5 23 12 2 6 2 4 8 5 0 49 5 2 31 16 0 28 2 8 17 6 3 2
Last week 5553 258 1079 435 2150 275 525 2931 1122 81 947 406 96 4113 44 127 369 52 107 20 13 173 203 32 308 93 9 28 62 239 148 66 87 24 48 72 30 34 405 37 19 398 152 5 229 34 45 134 56 44 33
Days since peak 12 42 7 88 12 64 17 56 61 40 38 53 52 56 51 30 40 56 26 33 47 62 55 62 51 47 54 18 41 40 17 26 17 61 32 64 69 47 55 40 47 73 31 53 70 17

Initial visual evaluation of forecasts of Deaths