it's a genuinely good use case too chalk this up as a W for the tradrats
-
-
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This Tweet is unavailable.
-
make a bunch of predictions at varying levels of confidence and see how many pan out eg, 80% of 80% confidence predictions should come true, if 90% do you're underconfident, if 60% you're overconfident
End of conversation
-
-
-
GFS or ECMWF?
-
i dont know hmm this could either be a genuine question or a clever joke
- Show replies
New conversation -
-
-
We pitched calibration to a very famous sports team and they too were very into the idea; simple concepts but useful results I guess
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Umm, hi. Long time follower, first time reply guy. Would you be able to say a little about the merhod(s) you employed? I’ve used bug tracker that gathered historical data on developer estimate vs actual and then created a probability distribution curve using
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Monte Carlo simulation to provide a confidence distribution curve and I’m curious to know how you produced your system. Thanks.
-
oh far more lazily i just kind of mentioned it offhand as an option for getting better at forecasts that inevitably involve a lot of guesswork, and the engineers wanted to know more so I walked them through it casually very informal
- Show replies
New conversation -
-
-
You should pitch them on momentum accounting their forecasting as well, then time-weight the results for maximum pedant-factor
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.