COVID-19 short-term forecasts Deaths 2020-07-18 Latin American Countries


Gnereal 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 Latin America 2020-07-18

ArgentinaBoliviaBrazilChileColombiaDominican RepublicEcuadorEl SalvadorGuatemalaHondurasMexicoPanamaPeru
Peak date --07-12 -- -- --04-1205-1007-02 --07-0206-24 --06-14
Peak daily increment 58 22 170 9 21 666 279
Days from 100 to peak 65 5 39 10 55 79 68
Days from peak/2 to peak 86 17 40 83 90 72 65
Last total 2220 2106 78772 8445 6516 971 5282 324 1443 891 38888 1071 12998
Last daily increment 42 57 921 98 228 29 32 15 0 34 578 33 199
Last week 375 299 6672 1466 1209 74 235 64 224 117 3882 162 1128
Days since peak 6 97 69 16 16 24 34

Deaths count forecast Latin America (bold red line in graphs) 2020-07-19 to 2020-07-25

DateArgentinaBoliviaBrazilChileColombiaDominican RepublicEcuadorEl SalvadorGuatemalaHondurasMexicoPanamaPeru
2020-07-18 2220 2106 78772 8445 6516 971 5282 324 1443 891 38888 1071 12998
2020-07-19 2291 2159 79470 8810 6633 983 5326 333 1481 912 39190 1102 13180
2020-07-20 2364 2209 80160 8990 6764 994 5365 342 1519 932 39630 1134 13360
2020-07-21 2439 2260 81440 9220 6881 1006 5403 351 1554 952 40400 1166 13530
2020-07-22 2516 2310 82580 9540 7026 1018 5441 359 1591 971 41020 1199 13710
2020-07-23 2596 2361 83790 9840 7183 1029 5479 368 1627 991 41660 1233 13880
2020-07-24 2678 2412 84950 10300 7391 1041 5517 376 1664 1011 42320 1268 14060
2020-07-25 2762 2463 85890 10590 7716 1053 5556 385 1702 1031 42860 1304 14240

Deaths count average forecast Latin America (bold black line in graphs) 2020-07-19 to 2020-07-25

DateArgentinaBoliviaBrazilChileColombiaDominican RepublicEcuadorEl SalvadorGuatemalaHondurasMexicoPanamaPeru
2020-07-18 2220 2106 78772 8445 6516 971 5282 324 1443 891 38888 1071 12998
2020-07-19 2281 2147 79400 8640 6745 979 5314 327 1478 900 39030 1094 13160
2020-07-20 2355 2195 80180 8850 6952 990 5348 332 1521 917 39430 1121 13350
2020-07-21 2431 2246 81400 9090 7160 1003 5390 339 1564 936 40230 1153 13550
2020-07-22 2510 2300 82540 9350 7405 1015 5424 347 1609 956 40870 1183 13750
2020-07-23 2589 2354 83700 9670 7703 1028 5465 356 1656 975 41530 1216 13950
2020-07-24 2673 2413 84840 10140 8008 1041 5506 364 1703 999 42190 1251 14170
2020-07-25 2756 2474 85900 10540 8362 1054 5549 373 1753 1023 42690 1289 14380

Deaths count scenario forecast (bold purple line in graphs) 2020-07-19 to 2020-07-27

DateArgentinaBoliviaBrazilChileColombiaDominican RepublicEcuadorEl SalvadorGuatemalaHondurasMexicoPanamaPeru
2020-07-18 2220 2106 78772 8445 6516 971 5282 324 1443 891 38888 1071 12998
2020-07-19 2293 2137 79600 8600 6686 977 5318 324 1495 901 39270 1097 13120
2020-07-20 2358 2170 80590 8900 6901 991 5349 332 1540 919 39750 1123 13250
2020-07-21 2425 2200 81430 9150 7114 1004 5382 340 1588 934 40150 1147 13390
2020-07-22 2489 2231 82390 9400 7340 1014 5411 348 1637 947 40530 1172 13520
2020-07-23 2555 2248 83370 9650 7529 1023 5436 358 1681 961 40890 1198 13620
2020-07-24 2613 2271 84260 9880 7737 1033 5456 369 1729 974 41230 1221 13720
2020-07-25 2680 2291 85100 10210 7909 1044 5479 378 1775 984 41520 1239 13820
2020-07-26 2741 2311 85920 10450 8074 1055 5496 389 1820 996 41860 1256 13890
2020-07-27 2798 2323 86710 10730 8208 1065 5514 395 1862 1007 42130 1270 13980

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