Publications by free range statistics - R
Test positivity rates and actual incidence and growth of diseases by @ellis2013nz
Test positivity rates and disease incidence OK, lets talk about test positivity rates ie the proportion of administered tests that conclude the subject has the disease being tested for. These are in the news, with COVID-19 test positivity being used as an informal indicator of COVID-19 incidence and the quality of the confirmed case data, for exa...
13586 sym 12 img
Incidence of COVID-19 in Texas after adjusting for test positivity by @ellis2013nz
Amidst controversy in several (perhaps many?) countries about the timing and pace of opening up from COVID-19 control measures, one small corner of the argument today related to why Texas is seeing record numbers of new cases of COVID-19 in the days after a range of opening up measures. In a thread on Twitter, @SeanTrende argued that the worrying...
3224 sym R (3368 sym/1 pcs) 4 img
Ordering bars within their clumps in a bar chart by @ellis2013nz
A pleasant diversion An unexpectedly pleasant diversion from the news during the COVID-19 pandemic has been the contemplation of a considerable range of data visualisations, from the truly excellent to the amusingly bad. After all, at the excellent end, the crisis has given us the Siouxsie Wiles and Toby Morris “flattening the curve” animated...
5848 sym R (3490 sym/2 pcs) 4 img
Forecasts for the 2020 New Zealand elections using R and Stan by @ellis2013nz
New New Zealand election forecasts I have finalised the first proper release version of my forecast for the New Zealand general election to be held on 19 September 2020. Here are the predicted probability distributions for number of seats for several of the realistic possible coalitions: As can be seen, it’s looking very good for Prime Ministe...
9256 sym 6 img
Estimating Covid-19 reproduction number with delays and right-truncation by @ellis2013nz
This great preprint recently came out from a team of Katelyn Gostic and others. It uses simulations to test various methods of estimating the effective reproduction number . If you are following the Covid-19 pandemic from a data angle at all, you will no doubt have come across the effective reproduction number and will know that it is an estimate...
18446 sym R (15082 sym/7 pcs) 14 img
Visualisation options to show growth in occupations in the Australian health industry by @ellis2013nz
Visualising growth in occupations in one industry A chart is doing the rounds purporting to show the number of administrators working in health care in the USA has grown much faster than the number of physicians – more than 3,000% growth from 1970 to 2009 for administrators (allegedly) compared to about 90% or so for physicians. I don’t much ...
7423 sym R (11786 sym/1 pcs) 14 img
Essentially random isn’t the same as actually random by @ellis2013nz
So an article/website is doing the rounds describing some country-level observational analysis of the relationship between hydroxychloroquine usage in the early stages of Covid-19 and deaths from that disease per million population. It purports to show a huge positive impact – reducing the chance of death by about 80% (with a relative risk rati...
24813 sym R (4169 sym/5 pcs) 10 img
Lines of best fit by @ellis2013nz
A couple of months ago my Twitter feed featured several mass-participation stern critiques laced with sneering and mockery (aka “Twitter pile-ons”) for some naive regression modelling in the form of lines of best fit through scatter plots. All of the examples I’ve seen deserved some degree of criticism, but not to the virulent degree that t...
16023 sym R (4621 sym/4 pcs) 12 img
Time series cross validation of effective reproduction number nowcasts by @ellis2013nz
For a few weeks now I’ve been publishing estimates of the time-varying effective reproduction number for Covid-19 in Victoria (and as bonuses, NSW and Australia as a whole). Reproduction number is the average number of people each infected person infects in turn. My modelling includes nowcasts up to “today” of that number (and forecasts for...
6961 sym 6 img
Mixture distributions and reporting times for Covid-19 deaths in Florida by @ellis2013nz
Caution / who should read this – although this blog post uses some data on reporting times for Covid-19 deaths in Florida, it’s not really about that issue at all. It’s a quick dip into how to model data that comes from a mixture of two or more different probability distributions. Mixture distributions I have thoughts about mixture distribu...
8028 sym R (11644 sym/7 pcs) 10 img