search

matthewdwood82 / ETR-Slack-Data-Science-Book-Club

Continuation of ETR Slack Book Club, focused on data science topics. General comms via slack/discord; book discussion via Github Issues & Google Meet.
1 stars 0 forks source link

Definitions 3.5/3.6 #6

Open jdsdog10 opened 1 year ago

jdsdog10 commented 1 year ago

The way they differentiate between def5 and def6 is interesting. The estimate for theta in the frequentist coverage is almost always based off of previously observed data, right?. Similarly, how would you figure out the bayesian coverage if you aren't using some type of prior distribution?

matthewdwood82 commented 1 year ago

I interpreted the difference between definitions as one of whether this datapoint is considered part of the sample or not for purpose of calculating the 95% confidence interval and determining if this datapoint falls within it.

Bayesians include this datapoint as information to inform that distribution when determining if it falls within the 95% CI. Frequentists do not include this datapoint in their calculation of the interval. Instead, they test whether the point falls within a 95% CI that does not include that point.