abcd_z comments on Applied Bayes' Theorem: Calculating the probability that she's over me. Could somebody check my work? TT_TT - Less Wrong

-9 Post author: abcd_z 19 July 2013 02:59AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (23)

You are viewing a single comment's thread. Show more comments above.

Comment author: abcd_z 19 July 2013 03:19:27AM 0 points [-]

And remember that the "probability that she's over me," should be probability that she got over you since the last time you checked.

Wouldn't that be redundant?

Comment author: Mestroyer 19 July 2013 03:32:07AM 0 points [-]

I don't know why it would be redundant. If she has a certain probability of getting over you every day, and you called her and interacted normally Yesterday, then for it to be 30%, would require that she had a 30% chance of getting over you in a day, before you knew anything about whether she answered your call or not.

Comment author: abcd_z 19 July 2013 03:38:38AM 0 points [-]

I don't trust my accuracy of measurement from the time we broke up until now. The relationship between us has been...uncertain, and her ignoring my calls is the first behavioral cue I can point to and say "Okay, that means there's a good chance she's over me."

Comment author: Mestroyer 19 July 2013 03:45:54AM 3 points [-]

Oh, I didn't know you broke up.

You explicitly broke up, and your prior for "she's over you" is only 30%?

Also, if she was breaking up, but maybe changing her mind, a 90% chance of answering your call seems too high, but I have no experience with these matters.

Comment author: abcd_z 19 July 2013 03:53:14AM 0 points [-]

Ah. Sorry I didn't make that clear.
Yeah, that particular prior is low at least partly because of my wishful thinking.

Not that it seems to have helped. =/