COVID-19 short-term forecasts Deaths 2020-09-24


General information

  • 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. 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.
  • A list of notes is below. The most recent note:
    [2020-07-01] Modified the short-term model to allow for (slowly changing) seasonality. Many countries show clear seasonality after the initial period, likely caused by institutional factors regarding data collection. This seasonality was also getting in the way of peak detection. As a consequence estimates of the peak date may have changed for countries with strong seasonality.
    Added forecasts of cumulative confirmed cases for lower tier local authorities of England. The data is available from 2020-07-02 including all tests (pillar one and two). Only authorities with more than 5 cases in the previous week are included.

Peak increase in estimated trend of Deaths in Europe 2020-09-24

UKEUATBEBGBSCZDEDKESFIFRGRHRHUIEITNLPLPTROSENOCH
Peak date04-1004-0604-0604-1508-2704-11 --04-1504-0403-3104-2104-08 -- --04-1904-2403-2904-0404-2504-1508-2504-1204-0704-06
Peak daily increment 964 3095 20 328 9 7 226 15 946 30 943 13 167 781 153 23 31 44 98 8 60
Days from 100 to peak 23 33 7 22 105 0 22 3 18 1 23 6 21 25 15 19 17 145 18 -2 14
Days from peak/2 to peak 17 23 16 21 153 22 22 15 14 20 18 26 22 20 16 31 23 151 21 20 19
Last total 41902 145518 783 9965 785 277 567 9436 645 31118 343 31333 366 261 709 1797 35781 6312 2369 1931 4591 5878 270 2061
Last daily increment 40 349 6 6 6 2 12 13 2 84 0 81 9 4 7 3 23 16 25 3 41 2 3 1
Last week 170 1739 20 28 32 2 72 50 10 623 4 248 39 17 40 5 113 39 99 37 231 13 3 16
Days since peak 167 171 171 162 28 166 162 173 177 156 169 158 153 179 173 152 162 30 165 170 171

Peak increase in estimated trend of Deaths 2020-09-24

AustraliaBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-COUS-CTUS-DCUS-DEUS-FLUS-GAUS-IAUS-IDUS-ILUS-INUS-KSUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NDUS-NEUS-NHUS-NJUS-NMUS-NVUS-NYUS-OHUS-OKUS-ORUS-PAUS-SCUS-SDUS-TNUS-TXUS-UTUS-VAUS-WAUS-WIUS-WV
Peak date09-0307-2105-0607-1708-2506-16 --07-2706-2407-2309-1405-2907-2204-1504-1309-1609-1507-2108-0604-2404-2004-2906-2308-0708-2205-2109-0905-1205-0209-2004-1404-1404-2504-2904-1405-28 --08-26 -- --05-2005-2906-2505-1308-1305-1808-2908-2407-2805-0507-28 --08-2007-2704-1704-2204-0604-1104-23
Peak daily increment 34 1065 168 785 329 1661 214 683 2947 158 171 323 125 2226 91 127 78 138 72 105 11 46 175 67 14 16 112 99 11 10 64 459 56 137 23 33 40 25 1566 9 18 3573 88 10 5 137 45 26 386 6 104 25 12 16
Days from 100 to peak 108 115 36 93 135 72 144 79 107 166 48 84 18 28 156 122 100 131 21 18 10 60 127 144 27 60 41 28 152 0 17 23 22 17 41 135 0 17 85 16 123 59 147 133 89 32 103 130 115 -45 7 15 2 -89
Days from peak/2 to peak 157 106 37 63 128 52 144 72 78 168 49 76 20 19 163 138 108 133 27 20 30 74 127 147 50 160 44 32 175 19 21 20 27 18 56 148 30 31 81 41 139 49 151 151 131 33 119 142 111 18 11 30 18 4
Last total 869 139808 9297 12469 24746 91149 10105 25015 75439 31870 5127 19867 16283 7785 202798 2506 1246 5557 15393 2033 4499 621 630 13795 6660 1303 457 8538 3322 621 1137 5241 9356 3897 6939 1988 1828 2874 3356 211 462 438 16091 859 1564 33080 4715 981 539 8054 3279 210 2282 15510 367 3111 2076 1265 328
Last daily increment 8 1703 3 124 0 1129 128 175 490 302 36 147 77 74 914 18 17 32 102 3 2 0 1 177 46 5 3 30 17 1 13 16 15 5 7 3 12 4 40 8 0 0 9 2 8 6 28 11 1 16 17 8 36 138 0 24 -1 6 6
Last week 25 4015 41 270 1081 5530 769 1063 2636 724 297 739 426 408 4228 78 73 108 500 24 7 2 10 570 274 39 19 127 52 27 36 69 93 38 65 38 119 82 149 27 20 0 30 18 40 14 107 42 18 133 102 15 115 559 5 164 44 27 28
Days since peak 21 65 141 69 30 100 59 92 63 10 118 64 162 164 8 9 65 49 153 157 148 93 48 33 126 15 135 145 4 163 163 152 148 163 119 29 127 118 91 134 42 129 26 31 58 142 58 35 59 160 155 171 166 154

Deaths count forecast Europe (bold red line in graphs) 2020-09-25 to 2020-10-01

DateUKEUATBEBGCZDEESFRGRHUITNLPLPTRO
2020-09-24 41902 145518 783 9965 785 567 9436 31118 31333 366 709 35781 6312 2369 1931 4591
2020-09-25 41940 145800 786 9968 794 583 9452 31210 31340 374 714 35800 6317 2369 1936 4629
2020-09-26 41990 146000 789 9970 795 592 9461 31290 31350 382 719 35820 6326 2382 1940 4665
2020-09-27 42020 146200 792 9972 799 604 9468 31370 31350 392 725 35840 6335 2393 1945 4702
2020-09-28 42060 146400 795 9974 804 617 9474 31450 31360 402 730 35850 6343 2401 1949 4738
2020-09-29 42110 146700 798 9977 809 632 9484 31530 31370 412 735 35870 6355 2418 1953 4774
2020-09-30 42150 146900 801 9979 817 641 9496 31600 31370 420 740 35880 6366 2440 1958 4810
2020-10-01 42170 147100 804 9981 823 655 9503 31680 31380 428 746 35900 6375 2459 1962 4846

Deaths count forecast (bold red line in graphs) 2020-09-25 to 2020-10-01

DateAustraliaBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-COUS-FLUS-GAUS-IAUS-ILUS-INUS-KSUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NDUS-NEUS-NJUS-NVUS-OHUS-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WIUS-WV
2020-09-24 869 139808 9297 12469 24746 91149 10105 25015 75439 31870 5127 19867 16283 7785 202798 2506 1246 5557 15393 2033 13795 6660 1303 8538 3322 621 1137 5241 9356 3897 6939 1988 1828 2874 3356 211 462 16091 1564 4715 981 8054 3279 2282 15510 3111 2076 1265 328
2020-09-25 881 140700 9302 12530 24930 92060 10250 25180 75890 32020 5176 19940 16350 7862 203800 2523 1258 5585 15530 2033 13930 6700 1317 8557 3336 628 1152 5286 9370 3903 6944 2000 1863 2899 3385 214 464 16100 1587 4761 989 8054 3322 2321 15680 3136 2081 1275 332
2020-09-26 892 141400 9308 12580 25100 92850 10400 25340 76290 32160 5225 20070 16420 7937 204500 2537 1270 5604 15630 2038 14000 6747 1318 8580 3345 635 1158 5286 9384 3908 6950 2010 1902 2913 3414 217 464 16110 1593 4771 995 8068 3334 2346 15770 3160 2087 1281 336
2020-09-27 903 141800 9313 12620 25270 93660 10550 25500 76490 32300 5273 20150 16490 8011 204700 2540 1282 5610 15630 2039 14000 6762 1319 8591 3348 642 1161 5312 9398 3914 6958 2014 1928 2916 3419 221 464 16110 1594 4772 998 8082 3345 2350 15810 3185 2094 1281 340
2020-09-28 913 142100 9318 12650 25440 94430 10710 25660 76660 32420 5321 20210 16560 8082 205000 2545 1293 5610 15670 2041 14020 6767 1331 8597 3353 650 1162 5320 9413 3919 6965 2017 1955 2919 3425 225 469 16120 1596 4782 999 8105 3357 2362 15850 3209 2100 1282 345
2020-09-29 924 143000 9322 12680 25600 94900 10900 25810 77280 32550 5369 20350 16630 8154 205900 2562 1305 5621 15790 2047 14130 6822 1340 8621 3364 657 1168 5335 9427 3924 6972 2024 1999 2946 3461 228 475 16130 1610 4809 1009 8117 3382 2388 15950 3233 2107 1288 349
2020-09-30 934 143200 9327 12700 25770 95690 11090 25970 77770 32670 5418 20490 16690 8226 206900 2588 1317 5646 15890 2053 14310 6887 1348 8645 3375 664 1175 5341 9442 3930 6979 2030 2061 2969 3494 232 479 16140 1620 4852 1018 8146 3402 2407 16080 3257 2113 1297 353
2020-10-01 945 144500 9332 12820 25940 96750 11270 26130 78210 32790 5466 20620 16760 8299 207800 2602 1329 5682 15990 2056 14480 6936 1352 8672 3389 671 1188 5356 9457 3935 6986 2036 2080 2980 3529 236 481 16150 1628 4879 1028 8159 3420 2439 16180 3281 2120 1302 357

Deaths count average forecast Europe (bold black line in graphs) 2020-09-25 to 2020-10-01

DateUKEUATBEBGCZDEESFRGRHUITNLPLPTRO
2020-09-24 41902 145518 783 9965 785 567 9436 31118 31333 366 709 35781 6312 2369 1931 4591
2020-09-25 41910 145700 784 9969 788 578 9440 31190 31350 368 714 35790 6313 2372 1935 4613
2020-09-26 41920 145900 785 9975 790 590 9443 31280 31380 371 720 35800 6314 2378 1940 4647
2020-09-27 41940 146100 786 9979 794 602 9446 31360 31410 378 726 35810 6316 2385 1945 4681
2020-09-28 41950 146300 788 9981 799 616 9448 31450 31440 385 733 35830 6317 2394 1950 4716
2020-09-29 41970 146500 789 9983 804 630 9454 31530 31480 390 739 35840 6321 2408 1955 4750
2020-09-30 41990 146700 791 9986 811 644 9461 31620 31510 395 746 35860 6324 2422 1960 4785
2020-10-01 42010 146900 793 9990 817 660 9465 31710 31540 403 752 35870 6327 2435 1965 4820

Deaths count average forecast (bold black line in graphs) 2020-09-25 to 2020-10-01

DateAustraliaBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-COUS-FLUS-GAUS-IAUS-ILUS-INUS-KSUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NDUS-NEUS-NJUS-NVUS-OHUS-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WIUS-WV
2020-09-24 869 139808 9297 12469 24746 91149 10105 25015 75439 31870 5127 19867 16283 7785 202798 2506 1246 5557 15393 2033 13795 6660 1303 8538 3322 621 1137 5241 9356 3897 6939 1988 1828 2874 3356 211 462 16091 1564 4715 981 8054 3279 2282 15510 3111 2076 1265 328
2020-09-25 872 140400 9301 12510 24830 91800 10250 25110 75670 31920 5178 19930 16310 7835 203400 2514 1255 5572 15480 2035 13890 6678 1314 8552 3333 627 1142 5260 9366 3900 6943 1996 1842 2885 3377 212 463 16100 1576 4747 985 8059 3300 2304 15600 3129 2080 1270 332
2020-09-26 875 141000 9306 12550 25060 92650 10420 25220 76040 32000 5241 20040 16360 7898 203800 2523 1267 5586 15580 2038 13950 6709 1316 8573 3343 634 1148 5268 9378 3904 6950 2006 1863 2899 3405 216 465 16100 1583 4758 990 8069 3315 2327 15690 3150 2081 1275 337
2020-09-27 879 141400 9311 12600 25220 93510 10570 25340 76180 32100 5305 20130 16410 7961 204000 2529 1280 5598 15580 2040 13960 6726 1320 8586 3345 641 1152 5289 9392 3909 6957 2010 1882 2903 3415 218 466 16100 1586 4759 993 8081 3328 2340 15740 3170 2083 1276 340
2020-09-28 883 141900 9317 12640 25370 94420 10740 25450 76330 32210 5369 20220 16470 8024 204200 2536 1294 5607 15610 2043 13980 6738 1330 8595 3349 649 1156 5301 9405 3914 6964 2013 1903 2907 3425 221 470 16110 1589 4766 995 8102 3342 2357 15780 3191 2088 1277 345
2020-09-29 887 142600 9323 12690 25490 95270 10930 25560 76940 32340 5434 20340 16530 8088 204700 2548 1308 5622 15730 2048 14080 6783 1338 8614 3361 658 1163 5318 9419 3920 6970 2021 1931 2931 3459 224 474 16110 1602 4792 1003 8114 3364 2383 15880 3211 2096 1283 349
2020-09-30 890 143300 9327 12730 25740 96170 11120 25680 77360 32460 5500 20460 16590 8151 205400 2565 1323 5641 15840 2051 14240 6821 1346 8639 3374 667 1171 5333 9433 3925 6977 2027 1962 2955 3493 229 477 16120 1613 4836 1010 8130 3388 2407 16010 3232 2101 1291 354
2020-10-01 894 144200 9333 12810 25930 97230 11300 25800 77610 32590 5566 20580 16650 8216 205900 2575 1338 5666 15940 2055 14400 6862 1351 8662 3383 677 1181 5352 9447 3930 6985 2036 1982 2973 3522 234 480 16120 1625 4869 1016 8143 3409 2433 16100 3252 2105 1295 358

Deaths count scenario forecast (bold purple line in graphs) 2020-09-25 to 2020-10-03

DateUKEUATBEBGCZDEESFRGRHUITNLPLPTROAustraliaBrazilCanadaChileColombiaIndiaIndonesiaIranMexicoPeruPhilippinesRussiaSouth AfricaTurkeyUSUS-ALUS-ARUS-AZUS-CAUS-COUS-FLUS-GAUS-IAUS-ILUS-INUS-KSUS-KYUS-LAUS-MAUS-MDUS-MIUS-MNUS-MOUS-MSUS-NCUS-NDUS-NEUS-NJUS-NVUS-OHUS-OKUS-PAUS-SCUS-TNUS-TXUS-VAUS-WAUS-WIUS-WV
2020-09-24 41902 145518 783 9965 785 567 9436 31118 31333 366 709 35781 6312 2369 1931 4591 869 139808 9297 12469 24746 91149 10105 25015 75439 31870 5127 19867 16283 7785 202798 2506 1246 5557 15393 2033 13795 6660 1303 8538 3322 621 1137 5241 9356 3897 6939 1988 1828 2874 3356 211 462 16091 1564 4715 981 8054 3279 2282 15510 3111 2076 1265 328
2020-09-25 41900 145700 783 9970 788 579 9438 31200 31350 369 715 35800 6312 2370 1935 4617 871 140300 9305 12510 25090 91930 10200 25150 75740 31890 5216 19920 16340 7842 203300 2507 1253 5565 15470 2035 13820 6674 1308 8556 3329 637 1142 5257 9365 3903 6944 1999 1834 2892 3372 211 465 16100 1574 4740 981 8073 3296 2300 15590 3130 2084 1270 332
2020-09-26 41920 146000 787 9980 792 595 9450 31360 31400 374 721 35820 6312 2387 1941 4652 873 140700 9314 12540 25220 92390 10320 25290 76110 31970 5272 20020 16400 7886 204100 2521 1267 5583 15530 2038 13910 6706 1315 8586 3340 646 1146 5272 9379 3911 6946 2005 1847 2906 3391 215 468 16100 1583 4761 987 8092 3312 2318 15660 3152 2092 1274 337
2020-09-27 41950 146200 789 9990 796 608 9462 31490 31440 378 727 35840 6315 2406 1946 4672 875 141100 9324 12580 25310 93000 10430 25420 76460 32040 5332 20120 16450 7940 204800 2528 1275 5599 15600 2041 13990 6733 1319 8615 3353 653 1150 5289 9392 3919 6950 2010 1858 2918 3403 219 470 16100 1591 4780 993 8117 3328 2331 15740 3178 2099 1279 340
2020-09-28 41950 146400 792 9990 799 622 9473 31600 31490 382 733 35870 6320 2423 1952 4700 877 141600 9335 12610 25450 93370 10560 25530 76790 32090 5380 20200 16500 7972 205500 2535 1285 5615 15660 2043 14060 6761 1322 8647 3368 661 1154 5304 9401 3927 6952 2015 1868 2929 3425 223 472 16110 1597 4799 998 8156 3343 2346 15810 3209 2105 1284 344
2020-09-29 41960 146500 794 10000 802 635 9484 31790 31520 388 739 35890 6324 2434 1958 4730 879 142100 9347 12640 25590 93900 10670 25620 77140 32160 5430 20280 16540 8021 206400 2546 1293 5633 15720 2045 14130 6784 1326 8686 3387 669 1158 5316 9410 3936 6954 2018 1884 2939 3448 227 474 16110 1604 4819 1002 8185 3354 2360 15880 3233 2112 1289 346
2020-09-30 41970 146600 796 10010 805 648 9501 31950 31550 393 744 35910 6327 2449 1963 4750 880 142400 9360 12670 25690 94270 10790 25710 77490 32210 5483 20330 16580 8052 207300 2551 1308 5648 15770 2048 14190 6806 1327 8722 3407 678 1162 5326 9416 3944 6955 2022 1896 2949 3467 232 475 16120 1610 4834 1006 8213 3369 2371 15950 3259 2118 1293 349
2020-10-01 41980 146800 798 10020 808 659 9516 32170 31590 396 749 35930 6330 2459 1968 4771 881 142900 9373 12690 25750 94430 10920 25780 77830 32260 5512 20400 16620 8078 208000 2557 1321 5662 15820 2049 14240 6826 1330 8764 3427 688 1166 5342 9421 3953 6957 2026 1906 2958 3481 236 476 16120 1615 4851 1010 8248 3382 2383 16010 3283 2123 1298 351
2020-10-02 41990 146900 800 10030 811 671 9527 32380 31610 401 755 35950 6331 2469 1973 4794 883 143400 9386 12720 25870 94640 11050 25840 78170 32330 5548 20450 16660 8116 208800 2562 1334 5674 15870 2049 14300 6847 1331 8806 3451 695 1171 5357 9428 3960 6959 2028 1916 2966 3495 241 476 16120 1619 4862 1012 8285 3393 2391 16090 3308 2126 1303 353
2020-10-03 42000 147000 803 10040 813 682 9541 32540 31640 406 758 35970 6332 2478 1977 4812 884 143800 9399 12750 25950 94840 11160 25900 78500 32380 5586 20490 16700 8147 209600 2567 1342 5685 15910 2051 14340 6864 1332 8847 3474 704 1177 5370 9434 3968 6962 2031 1928 2974 3508 245 476 16130 1623 4876 1015 8316 3401 2403 16150 3329 2131 1307 355

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.

Recent changes and notes

[2020-07-01] Modified the short-term model to allow for (slowly changing) seasonality. Many countries show clear seasonality after the initial period, likely caused by institutional factors regarding data collection. This seasonality was also getting in the way of peak detection. As a consequence estimates of the peak date may have changed for countries with strong seasonality.
Added forecasts of cumulative confirmed cases for lower tier local authorities of England. The data is available from 2020-07-02 including all tests (pillar one and two). Only authorities with more than 5 cases in the previous week are included.
[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 streamed here.
[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.
[2020-06-06] Removed Brazil from yesterday's forecasts (only; last observation 2020-06-05).
[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.
[2020-05-20] Problem with UK confirmed cases: negative daily count. This makes the forecasts temporarily unreliable.
Updated the second paper.
[2020-05-18] Minor fixes to the improved version of scenario forecasting, backported to 2020-05-13.
[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.
[2020-05-06] 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.
[2020-04-29] 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.
[2020-04-27] 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.
[2020-04-24] A summary of our work on short-term COVID-19 forecasting appeared as a voxeu.
[2020-04-17] Bird and Nielsen look into nowcasting death counts in England.
[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).
[2020-04-10] Updated documentation with better description of short-term estimates and peak determination.
[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.
[2020-04-08] Minor correction to peak estimates. Added table with scenario forecasts.
[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.
[2020-04-02] Now including more US States, based on New York Times data.
[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).
[2020-03-26] Scenario forecasts that are based on what happened in China earlier this year, only for Italy.
[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.

Initial visual evaluation of forecasts of Deaths