Are there differential models of NPV as affected by current service instance quality? Formula for how one bad coffee changes my LTV to cafe?
-
-
Replying to @vgr
You should view it as a bayesian value of information question; how much does it change your posterior estimate of value in the future?
1 reply 0 retweets 0 likes -
Replying to @davidmanheim @vgr
And no, I don't think that everything is about Bayesian VoI just because that's my dissertation topic. . . . Just most things. :)
1 reply 0 retweets 1 like -
Replying to @davidmanheim
actually I think in this case, it's about narrative rewriting...what *account* customer gives themselves of expanded data
1 reply 0 retweets 3 likes -
Replying to @vgr @davidmanheim
you can still apply Bayesian reasoning on the set of narratives in play (this was my postdoc stuff)
2 replies 1 retweet 2 likes -
Replying to @vgr
But isn't applying a bayesian model to the narrative (= utility, value) isomorphic to applying it to the data, then calculating utility
2 replies 0 retweets 1 like -
Replying to @davidmanheim
re: isomorphism... I think yes for sufficiently small data, with exhaustive, mutually exclusive coverage by narratives.
1 reply 0 retweets 2 likes -
Replying to @davidmanheim @vgr
I was considering if the narrative helps precisely when the world is too large to compute by giving simpler model
1 reply 0 retweets 1 like
*hands cigar to da man* ... Yup precisely. They carve out the subset of data you want to consider more carefully
-
-
Replying to @vgr @davidmanheim
interestingly, this would suggest post-hoc narratives from those with better models can help more than ex-ante ones
2 replies 0 retweets 1 like -
as it allows the expert to encode which aspects of the instantiation are more revealing of the underlying mechanism
1 reply 0 retweets 1 like - 8 more replies
New conversation -
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.