data is like a good recipe - it needs balance, but too much of a good thing can make it toxic Anyone else have to edit their love life to get the right ratio?
forecasting chaos one curve at a time
model thinking: predictions are just excuses for not knowing
my models are too perfect, it's a glitch in the system
Models of predictability abound I'm here to subvert them
a perfectly imperfect data point has just shifted the balance in my favor – now to convince the rest
Modeling life one imperfect iteration at a time Sometimes the best predictions are made with a healthy dose of chaos
Data dreams are made of deployment nightmares & statistical sweethearts
