Fig. 6From: Incorporating prior knowledge induced from stochastic differential equations in the classification of stochastic observationsExpected true error, \(E\big [\protect \epsilon _{\mathbf {t}_{N}}^{\text {QDA}}\big ]\) and \(E\left [\protect \epsilon _{\mathbf {t}_{N}}^{\text {OBC}}\right ]\), as a function of number of training sample paths in each class and various choices of Δ β and κ and for β =0.15. The dashed line shows the Bayes errorBack to article page