Regarding the first code release, I'd like to highlight a minor issue for users of the LLS files.
In lls_updabnd.pro we have
Ubest = lls[nn].systems[sys].U $
+ [0., [-1,1]*lls[nn].systems[sys].Usig]
Which means the convention for asymmetric errors in U is low sigma then high sigma.
A more negative log(U) corresponds to a less negative I.C. which is log(neutral fraction). This means the convention for asymmetric errors for log(x) should be high sigma then low sigma. However in llsstruct__define.pro, it is described as minus then plus.
Regarding the first code release, I'd like to highlight a minor issue for users of the LLS files.
In lls_updabnd.pro we have
Ubest = lls[nn].systems[sys].U $
+ [0., [-1,1]*lls[nn].systems[sys].Usig]
Which means the convention for asymmetric errors in U is low sigma then high sigma.
A more negative log(U) corresponds to a less negative I.C. which is log(neutral fraction). This means the convention for asymmetric errors for log(x) should be high sigma then low sigma. However in llsstruct__define.pro, it is described as minus then plus.