meeting 2023 10 12 - JacobPilawa/TriaxSchwarzschild_wiki_5 GitHub Wiki
Context
Here are just some plots in case we need them for discussion for the meeting -- these are all the cornerplots from the dynesty runs, as well as the individual chi2 vs parameters
Cutoffs for Chi2 <1 Case
Diagnostic information on creating the confidence intervals for the delta chi2 < 1 cases
* We saw that some of the Chi2<1 regions of the 1d panels are quite sparse, and imposing this cut leaves some models with only 1, 2, or 3 models in this interval
* I went ahead and did the following:
* first, impose the chi2 of 1 cut and get intervals for which there are data
* if the resulting uncertainty is one-sided, use the one-sided error on both sides
* if there are no other models, gradually increase the cutoff until a second model lands in the interval, at which point treat it as a second interval