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
This is because pcubature starts with a low order rule and samples the integrand at the three points 0, π and 2π for which sin^2 is identically 0. Similarly, when the integrand is cos^2 it returns 2π because the integrand at those points is 1. (Note that hcubature gives the correct result in both cases.)
Currently there is no way of specifying a higher order rule: can this be supported?
The text was updated successfully, but these errors were encountered:
pcubature
gives the wrong answer when integratingsin(x)^2
in the interval[0,2π]
: it should be π, it returns instead 0 (thread in discourse)This is because
pcubature
starts with a low order rule and samples the integrand at the three points 0, π and 2π for whichsin^2
is identically 0. Similarly, when the integrand iscos^2
it returns 2π because the integrand at those points is 1. (Note thathcubature
gives the correct result in both cases.)Currently there is no way of specifying a higher order rule: can this be supported?
The text was updated successfully, but these errors were encountered: