You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
That said, I am not sure that improving the precision of this function is the right way to tackle the problem. Apparently log2mexp is used in two places: the log(c)cdf of the Laplace function, and in tdistinvlogcdf in StatsFuns.
I am pretty sure that the first application at least should be handled as log1mexp(-z-loghalf), before rearranging the formula to use log2mexp. I am still looking at the second one.
Neither StatsFuns (JuliaStats/StatsFuns.jl#244) nor Distributions (JuliaStats/Distributions.jl#2115) uses LogExpFunctions.log2mexp anymore. StatsFuns and Distributions seemingly were the main (only?) users of log2mexp which raises the question: Is it worth keeping this function and the complexity required for numerically (reasonably) accurate evaluation in LogExpFunctions?
Activity