Top posts from 2021

1. From 2020:, the different ways the term “weights” is used in statistics and when it matters. See also, when is and isn’t it ok to just subset a survey data set by dropping records
2. From 2019: What have I got against the Shapiro-Wilk test?
3. From May: Why causal models are relevant to prediction: because they are relevant to generalisation. See also, mushrooms as an example
4. From February: Co-linearity. “Collinearity diagnostics aren’t much help, because they don’t tell you whether you’re interested in ${\beta }_X$ or ${\gamma }_X$. What they tell you is that if you’re interested in ${\gamma }_X$ it sucks to be you, because the data don’t provide much information about ${\gamma }_X$”
5. The April 1 post. You can do matrix multiplication with joins. This is not completely a joke
6. From August: Pictures of code are not code.

And one that didn’t get popular but that I liked: Data is the new oil