I recently saw this calmly frustrated tweet by John Myles White:
It’s odd how much of my life these days is spent fitting models I know are false to data that I believe to be inadequate.
It was a cry for help and I knew he needed some kind, guiding words to help him through his difficult time. Since Bayesianism is a religion, I replied with a Bayesian version of the Serenity Prayer.
Thankfully, I wasn’t the only one who found it amusing, so I thought I’d post an extended version here:
Bayesian Serenity Prayer
Bayes,
grant me the serenity
to accept the models are wrong;
the courage to clean the data;
and wisdom to choose a prior.Updating one datum at a time;
Sampling one model at a time;
Accepting iteration as the pathway to inference;
Taking, as He did, this uncertain world
as it might be, not as I would have it;
Trusting that He will make all things right
if I compute with His rule;
So that I may be approximately correct with this prediction
and exactly correct with His
Forever in the limit.
Amen.
Also: Happy New Year! I’m hoping to get back to some more serious blogging once a number of looming paper deadlines have passed.
Mark Reid January 24, 2011 Canberra, Australia