COVID-19 short-term forecasts Deaths 2020-06-22


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 The documentation that is provided is still in progress and not peer reviewed.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) 2020-06-23 to 2020-06-29

DateUKEUBEBGDEESFRITNLPLROSE
2020-06-22 42647 131997 9696 207 8899 28324 29595 34657 6090 1359 1523 5122
2020-06-23 42700 132000 9700 210 8910 28400 29600 34700 6090 1370 1540 5140
2020-06-24 42900 133000 9710 220 8920 28500 29600 34700 6100 1380 1550 5160
2020-06-25 43000 133000 9710 220 8930 28700 29700 34800 6100 1390 1570 5190
2020-06-26 43100 133000 9720 230 8950 28900 29700 34800 6110 1410 1580 5210
2020-06-27 43200 134000 9730 230 8960 29100 29700 34900 6110 1420 1600 5240
2020-06-28 43400 134000 9740 240 8970 29200 29800 34900 6120 1430 1610 5260
2020-06-29 43500 134000 9740 240 8990 29400 29800 35000 6120 1450 1630 5290

Deaths count average forecast (bold black line in graphs) 2020-06-23 to 2020-06-29

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-COUS-CTUS-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
2020-06-22 51271 8494 4502 2426 14011 2500 9742 22584 8223 1177 8196 1991 4974 120402 831 227 1350 5566 1651 4263 3173 2606 686 6671 2363 526 3004 7863 3052 6024 1385 939 978 1278 12974 469 489 31176 2704 6426 659 523 2207 1620 1276 745
2020-06-23 52000 8520 4640 2490 14300 2540 9800 23300 8310 1180 8310 2040 4990 121000 840 230 1360 5610 1660 4270 3200 2620 690 6700 2380 530 3020 7890 3070 6030 1390 940 980 1280 13000 470 490 31200 2720 6440 670 530 2220 1630 1280 750
2020-06-24 52900 8560 4800 2580 14600 2580 9900 24100 8400 1190 8450 2100 5010 121000 840 240 1380 5670 1660 4280 3220 2630 690 6740 2390 530 3030 7920 3080 6050 1400 950 990 1290 13100 480 490 31300 2730 6460 670 530 2250 1640 1290 760
2020-06-25 53800 8590 4970 2700 15000 2620 10000 25000 8510 1200 8600 2170 5030 122000 850 240 1400 5720 1670 4290 3250 2650 700 6780 2400 530 3040 7950 3100 6060 1410 960 990 1290 13100 480 500 31300 2750 6490 680 540 2270 1640 1300 760
2020-06-26 54700 8630 5150 2820 15300 2670 10100 26000 8640 1210 8740 2230 5050 123000 860 250 1420 5780 1670 4300 3280 2670 700 6820 2420 540 3050 7980 3120 6070 1430 970 990 1300 13200 480 500 31300 2770 6510 690 550 2290 1650 1300 770
2020-06-27 55600 8670 5330 2950 15700 2710 10200 27000 8770 1220 8890 2290 5070 123000 870 260 1440 5830 1680 4310 3310 2690 700 6870 2430 540 3060 8010 3140 6080 1440 980 1000 1310 13200 490 500 31400 2790 6540 690 550 2310 1660 1310 770
2020-06-28 56600 8700 5530 3080 16100 2760 10300 28000 8900 1240 9050 2360 5090 124000 870 260 1460 5890 1680 4320 3340 2710 700 6910 2450 540 3080 8040 3150 6090 1450 990 1000 1320 13300 490 510 31400 2800 6570 700 560 2340 1670 1320 780
2020-06-29 57500 8740 5730 3220 16500 2800 10400 29000 9060 1250 9200 2430 5100 124000 880 270 1480 5940 1690 4330 3370 2720 710 6950 2460 540 3090 8070 3170 6100 1460 1000 1010 1330 13400 500 510 31500 2820 6590 710 570 2360 1680 1330 790

Deaths count forecast Europe (bold red line in graphs) 2020-06-23 to 2020-06-29

DateUKEUBEBGDEESFRITNLPLROSE
2020-06-22 42647 131997 9696 207 8899 28324 29595 34657 6090 1359 1523 5122
2020-06-23 42800 132000 9700 210 8910 28400 29600 34700 6100 1370 1530 5150
2020-06-24 42900 133000 9710 220 8920 28400 29600 34700 6100 1370 1550 5170
2020-06-25 43000 133000 9710 230 8930 28400 29700 34800 6110 1380 1560 5190
2020-06-26 43000 133000 9720 230 8940 28500 29700 34800 6110 1390 1570 5210
2020-06-27 43100 133000 9720 240 8950 28500 29700 34900 6120 1390 1580 5230
2020-06-28 43200 134000 9730 240 8960 28600 29800 34900 6120 1400 1590 5250
2020-06-29 43300 134000 9740 250 8980 28600 29800 35000 6130 1410 1600 5270

Deaths count forecast (bold red line in graphs) 2020-06-23 to 2020-06-29

DateBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-COUS-CTUS-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
2020-06-22 51271 8494 4502 2426 14011 2500 9742 22584 8223 1177 8196 1991 4974 120402 831 227 1350 5566 1651 4263 3173 2606 686 6671 2363 526 3004 7863 3052 6024 1385 939 978 1278 12974 469 489 31176 2704 6426 659 523 2207 1620 1276 745
2020-06-23 51900 8530 4630 2510 14300 2540 9800 23400 8350 1190 8290 2040 5000 121000 840 230 1360 5610 1660 4270 3190 2620 690 6710 2380 530 3020 7900 3070 6040 1390 950 990 1290 13000 470 490 31200 2720 6440 670 530 2230 1630 1280 750
2020-06-24 52500 8560 4750 2600 14600 2580 9900 24300 8530 1200 8390 2090 5020 121000 840 230 1370 5660 1660 4280 3210 2630 690 6740 2390 530 3030 7920 3080 6050 1400 950 1010 1310 13000 470 490 31300 2730 6460 670 530 2250 1640 1290 750
2020-06-25 53000 8590 4880 2710 14900 2610 10000 25200 8720 1210 8490 2150 5040 122000 850 230 1380 5710 1670 4280 3230 2640 690 6770 2400 530 3040 7950 3100 6060 1410 960 1020 1330 13100 480 500 31300 2740 6480 680 540 2260 1640 1300 760
2020-06-26 53600 8620 5020 2810 15200 2650 10100 26100 8910 1220 8590 2210 5060 122000 850 240 1390 5750 1670 4290 3250 2650 700 6810 2420 530 3050 7980 3110 6070 1420 970 1030 1350 13100 480 500 31300 2750 6500 690 540 2280 1650 1310 760
2020-06-27 54200 8650 5150 2920 15500 2690 10300 27100 9100 1230 8690 2260 5080 123000 860 240 1400 5800 1670 4290 3270 2660 700 6840 2430 540 3060 8000 3130 6080 1420 970 1050 1370 13100 480 500 31400 2770 6520 700 550 2300 1660 1310 760
2020-06-28 54800 8670 5290 3040 15800 2730 10400 28100 9300 1240 8790 2320 5100 123000 860 240 1410 5850 1680 4300 3290 2670 700 6870 2440 540 3080 8030 3140 6090 1430 980 1060 1390 13100 480 500 31400 2780 6530 700 550 2320 1670 1320 770
2020-06-29 55300 8700 5440 3160 16100 2770 10500 29200 9500 1260 8900 2390 5120 124000 870 240 1420 5890 1680 4310 3310 2680 700 6910 2450 540 3090 8060 3160 6100 1440 990 1070 1410 13100 490 510 31400 2790 6550 710 560 2340 1670 1330 770

Deaths count scenario forecast (bold purple line in graphs) 2020-06-23 to 2020-07-01

DateUKEUBEBGDEESFRITNLPLROSEBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-COUS-CTUS-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
2020-06-22 42647 131997 9696 207 8899 28324 29595 34657 6090 1359 1523 5122 51271 8494 4502 2426 14011 2500 9742 22584 8223 1177 8196 1991 4974 120402 831 227 1350 5566 1651 4263 3173 2606 686 6671 2363 526 3004 7863 3052 6024 1385 939 978 1278 12974 469 489 31176 2704 6426 659 523 2207 1620 1276 745
2020-06-23 42800 132000 9710 210 8920 28600 29600 34700 6100 1370 1540 5160 52200 8530 4640 2550 14400 2540 9800 23100 8230 1190 8320 2030 4990 121000 840 230 1370 5610 1660 4270 3200 2630 690 6710 2380 530 3020 7900 3070 6040 1400 950 980 1280 13000 470 490 31200 2720 6470 670 530 2230 1630 1280 750
2020-06-24 42900 133000 9710 220 8930 28700 29600 34800 6100 1390 1550 5180 53100 8570 4750 2670 14800 2580 9900 23800 8290 1200 8430 2080 5010 121000 840 240 1380 5650 1660 4280 3230 2640 690 6740 2390 530 3030 7920 3080 6050 1410 960 980 1290 13000 470 500 31300 2730 6480 670 530 2240 1630 1290 760
2020-06-25 43000 133000 9720 220 8940 28900 29700 34800 6110 1400 1560 5200 54100 8590 4860 2780 15300 2610 10000 24400 8360 1210 8560 2130 5030 121000 850 250 1390 5680 1670 4280 3250 2650 700 6770 2400 530 3040 7950 3090 6060 1410 970 980 1300 13000 480 500 31300 2740 6490 680 540 2260 1640 1290 760
2020-06-26 43100 133000 9720 230 8950 29100 29700 34800 6110 1410 1570 5220 55200 8620 4960 2910 15600 2640 10200 25100 8440 1220 8660 2170 5050 122000 860 250 1390 5710 1670 4280 3270 2660 700 6800 2410 540 3050 7970 3110 6070 1420 980 980 1320 13100 480 500 31300 2750 6510 680 540 2270 1640 1300 770
2020-06-27 43200 133000 9730 230 8960 29300 29700 34900 6110 1420 1580 5230 56200 8650 5010 3030 16100 2680 10300 25700 8550 1230 8760 2210 5060 122000 860 260 1400 5740 1670 4290 3290 2670 700 6830 2420 540 3060 7990 3120 6070 1430 980 990 1320 13100 480 500 31400 2760 6530 690 550 2290 1650 1300 770
2020-06-28 43200 133000 9730 240 8960 29400 29700 34900 6120 1420 1590 5250 57400 8670 5090 3160 16500 2710 10400 26300 8550 1240 8850 2250 5080 122000 870 270 1400 5760 1680 4290 3310 2680 700 6850 2440 540 3070 8020 3130 6080 1430 990 990 1320 13100 480 500 31400 2770 6550 700 550 2310 1650 1310 780
2020-06-29 43300 133000 9730 240 8970 29500 29700 34900 6120 1430 1610 5260 58700 8690 5160 3280 16800 2740 10400 26900 8550 1240 8930 2290 5090 123000 870 270 1410 5780 1680 4300 3320 2690 710 6870 2450 540 3080 8030 3140 6090 1440 990 1000 1320 13100 480 510 31500 2780 6560 700 550 2320 1650 1310 780
2020-06-30 43400 133000 9740 250 8980 29500 29800 35000 6130 1430 1620 5270 59600 8710 5220 3390 17100 2770 10500 27200 8550 1250 9010 2330 5100 123000 880 280 1410 5800 1680 4310 3340 2700 710 6890 2460 540 3090 8050 3150 6090 1440 1000 1000 1320 13100 480 510 31500 2790 6580 710 550 2330 1660 1310 780
2020-07-01 43500 133000 9740 250 8990 29700 29800 35000 6130 1440 1620 5280 60800 8740 5280 3500 17400 2790 10600 27800 8550 1250 9090 2360 5120 123000 880 280 1420 5820 1690 4310 3350 2710 710 6910 2460 540 3090 8070 3160 6100 1450 1000 1010 1320 13200 490 510 31500 2790 6590 710 560 2340 1660 1320 790

Peak increase in estimated trend of Deaths in Europe 2020-06-22

UKEUATBEBGBSCZDEDKESFIFRHUIEITNLPLPTROSENOCH
Peak date04-0904-0704-0704-10 --04-1104-0904-1504-0403-3004-2104-0704-1904-2403-2804-0704-2404-1105-0104-2204-0804-04
Peak daily increment 1019 3232 22 349 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 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 22 14 22 14 13 24 16 26 25 19 18 28 19 37 27 21 17
Last total 42647 131997 690 9696 207 255 336 8899 602 28324 327 29595 572 1717 34657 6090 1359 1534 1523 5122 248 1956
Last daily increment 15 93 0 0 8 0 0 4 2 1 1 21 2 2 23 0 3 4 11 11 4 0
Last week 678 2119 9 33 26 0 5 79 4 1188 1 111 7 8 252 20 87 12 86 183 6 2
Days since peak 74 76 76 73 72 74 68 79 84 62 76 64 59 86 76 59 72 52 61 75 79

Peak increase in estimated trend of Deaths 2020-06-22

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-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WI
Peak date06-0405-0406-08 --06-1606-1803-19 -- --04-1205-2906-1504-2004-1505-06 --05-0804-2304-2404-2004-2505-1605-0604-2005-2005-1304-2904-1404-1604-1404-2704-2904-2205-2805-0505-06 --05-2005-2904-2905-1204-1204-0704-2904-2105-0504-2904-0605-1504-2204-0605-29
Peak daily increment 1097 179 302 975 50 147 27 184 71 125 2272 17 37 87 51 115 12 11 54 46 16 127 50 7 11 75 221 63 151 27 19 22 23 15 300 10 10 1068 54 9 170 18 10 39 47 31 14
Days from 100 to peak 68 33 53 72 81 14 11 48 47 23 30 23 26 26 21 18 5 22 35 20 26 43 25 -6 1 18 26 22 25 41 26 23 0 16 29 15 0 18 25 8 32 14 -5 41 7 14 51
Days from peak/2 to peak 59 34 47 59 90 16 25 48 61 25 21 37 37 30 28 20 26 47 41 25 48 44 32 18 21 21 26 26 26 55 39 38 41 31 28 40 18 13 31 27 33 34 8 50 14 30 65
Last total 51271 8494 4502 2426 14011 2500 9742 22584 8223 1177 8196 1991 4974 120402 831 227 1350 5566 1651 4263 535 435 3173 2606 686 6671 2363 257 526 3004 7863 3052 6024 1385 939 978 1278 248 339 12974 469 489 31176 2704 369 6426 659 523 2207 1620 1276 745
Last daily increment 680 12 23 73 312 35 119 759 362 8 95 61 24 425 2 2 2 51 4 3 2 0 12 4 0 24 13 0 0 11 16 14 7 4 4 40 55 5 0 35 0 2 51 4 0 3 6 5 16 9 6 1
Last week 6030 223 1119 618 2108 269 677 4274 1167 74 922 366 132 3439 52 39 123 366 34 53 15 11 180 116 21 274 98 11 21 74 208 115 63 72 77 63 109 18 13 182 22 22 267 105 6 150 52 37 158 50 45 42
Days since peak 18 49 14 6 4 95 71 24 7 63 68 47 45 60 59 63 58 37 47 63 33 40 54 69 67 69 56 54 61 25 48 47 33 24 54 41 71 76 54 62 48 54 77 38 61 77 24

Initial visual evaluation of forecasts of Deaths