Okay, I’ll bite. > If you want to convert people into loving linear models (and you should), we need to make sure that they learn the difference between 'linea

Okay, I’ll bite. > If you want to convert people into loving linear models (and you should), we need to make sure that they learn the difference between 'linear models' and 'linear models fit using OLS' Help me understand the pitch. What linear models are you referring to here that aren’t estimated with OLS? How should I wrap my head around having more parameters than observations?