[Machine learning] emphasizes approximating non-linear conditional-mean functions in highly-flexible non-parametric fashion. That turns out to be doubly unnecessary in econometric [time series]: There’s just not much conditional-mean non-linearity to worry about, and when there occasionally is, it’s typically of a highly-specialized nature best approximated in highly-specialized (tightly-parametric) fashion.
John Myles White, “What is an Interaction Effect?”:
The absence of interactions has nothing to do with linearity: it’s driven instead by a form of additivity.
After working with git, Dropbox stresses me out.
“This Book Left Me in Tears”, by Bill Gates