What a sensational title about a 14% correlation on a proxy variable! I wish I could sometimes turn into a journalist and sell my model with 30% MAPE (mean abso

What a sensational title about a 14% correlation on a proxy variable! I wish I could sometimes turn into a journalist and sell my model with 30% MAPE (mean absolute percentage error) to management like this. This is why popular medical press should not be read at all, because it announces another cure of cancer or of aids every day, yet those are just preliminary in-silico researches (simulation of chemical reaction), 90% of which fail at in-vitro stage, another 9% at in-vivo.