# Missing $inserted, any ideas where? I have been staring at the screen for what seems like hours trying to understand this error. I cannot see why it's giving me an error message. It's more than likely that I have been staring at the screen too long to see it but here is the code: Then clearly$P((q_a < {\theta} < q_b)|y) = 1 – {\alpha}$; so our confidence that${\theta}$lies in$(q_a, q_b)$is simply$100 {\times} (1-{\alpha})%$. This interval is a$100 {\times}
(1-{\alpha})%$credible set (or simply Bayesian confidence interval) for${\theta}$. This interval is relatively straightforward to calculate and has a direct interpretation that the frequentist interval does not. This interval is sometimes called the equal tail credible set for the simple reason that it is obtained by removing an equal amount of support$(\frac{{\alpha}}{2})$off both sides of$p({\theta}|y)$. Note that for symmetric unimodal posteriors, this equal tail interval will be symmetric about this mode If anyone can see where the error is I would really appreciate it. The compiler says it is in the line "This interval ....." but I can't see the issue - ## migrated from stackoverflow.comJan 11 '14 at 15:59 This question came from our site for professional and enthusiast programmers. Looks like your problem really is the same: you have underscores in your math blocks, you need to escape them. – Barmar Jan 10 '14 at 2:25 @Barmar: Underscores in math mode is acceptable and denotes a subscript. The problem is surely related to the use of the comment character %, which causes the math block to not be closed properly. You need to escape the percentage sign using \%. – Werner Jan 10 '14 at 3:24 ## 1 Answer Here:$100 {\times} (1-{\alpha})%$. This interval is a$100 {\times}
(1-{\alpha})%$The next$'s are then misplaced.

-