Exploration of COVID infection data in Italy: how high really was the first wave?

Official data for the COVID epidemic in Italy are availabe at https://github.com/pcm-dpc. The number of known active cases over time show a first wave of infections over March-June and a second one, much higher, started in October 2020 and still ongoing:

FIGURE 1. Number of known active COVID cases per day in Italy throughout 2020.
Continue reading

One hundred routes: my climbing in 2019 in numbers and pictures.

The year 2019 was the centenary of the foundation of the KwaZulu-Natal branch of the Mountain Club of South Africa (hereafter MCSA-KZN). To celebrate the occurrence, my climbing partner Neil came up with a challenge: to climb 100 different routes within the year. Sixteen people took up the challenge; of those, six of them reached the targeted number. Some would claim that climbing is all about the experience but, deep down, we all know that it is all about the numbers. Here are some.

Flashing “One step from the Earth” (17 trad), Morgan Bay. Gavin Raubenheimer at the belay. Picture by Anthony van Tonder.
Continue reading

ANOVA, model selection, and pairwise contrasts among treatments using R.

Some time ago I wrote about how to fit a linear model and interpret its summary table in R. At the time I used an example in which the response variable depended on two explanatory variables and on their interaction. It was a rather specific article, in which I overlooked some essential steps in the process of selection and interpretation of a statistical model. In the present article and in the relative worked examples I will expand the topic a bit, explaining 1) how to select the most parsimonious model relatively to a dataset using function anova() and 2) how to use summary(), together with relevel(), for testing for significant differences between pairs of experimental treatments (R script here). Continue reading