Country coronavirus country data revamped Archived Message Posted by walter on May 13, 2020, 2:27 am Changes to data format: 1) The two tables are combined into one so the gradients and all the other country information can be seen alongside the trend. 2) Instead of showing the trend for 8 selected countries, the top 28 countries (in terms of total cases) are shown. 3) I've switched to a 7-day cycle instead of 6, to give a longer time frame and to remove the effect of day-of-week fluctuations (eg UK figures are low on Mondays). So, the cases and deaths gradients are now the last week's total divided by the previous week's total. The present root product (shown in bold) is the same as before (i.e. the square root of the product of the cases and deaths gradients) with the same interpretation (>1='exponential' zone). The previous six root products are shown alongside. (Source: worldometer figures) Based on these figures the UK deaths may be set to come down; the deaths gradient (~0.6) has been falling anyway and the cases gradient has survived the govt ramping up the testing. Brazil still a worry, a 200m population and not yet got a handle on cases or deaths; Mexico(130m popn) may be following. Ecuador's minus root product is due to total reported deaths 'falling', must have been revised figures; the same happened in France. Another worrying aspect is that the trends in many countries near the top of the table - Germany, Turkey, France, Italy, Spain, Iran (total population of 400m), having been falling steadily, are now seemingly back on the rise, possibly reflecting lockdown easing. Sweden, Portugal, Belgium may be in the same boat. Message Thread: Country coronavirus country data revamped - walter May 13, 2020, 2:27 am Thanks Walter. How do you estimate reliability of the raw numbers on which you're working? As you - Rhisiart Gwilym May 13, 2020, 9:51 am Re: Thanks Walter. How do you estimate reliability of the raw numbers on which you're working? As you - walter May 13, 2020, 12:31 pm   Searching