Author Correction: Revisiting the concept of a symmetric index of agreement for continuous datasets

Correction to: Scientific Reports https://doi.org/10.1038/srep19401, published online 14 January 2016

This Article contains an error in Equation 15, where the $\kappa$ in the denominator is not multiplied by $n^{-1}$.

$$\lambda =1-\frac{n^{-1}{\mathrm{\Sigma }}_{i=1}^n(X_i-Y_i{)}^2}{{\sigma }_X^2+{\sigma }_Y^2+(\overline{X}-\overline{Y}{)}^2+\kappa }$$

should read:

$$\lambda =1-\frac{n^{-1}{\mathrm{\Sigma }}_{i=1}^n(X_i-Y_i{)}^2}{{\sigma }_X^2+{\sigma }_Y^2+(\overline{X}-\overline{Y}{)}^2+n^{-1}\kappa }$$

An extra note of caution is warranted regarding the variances ${\sigma }_X^2$ and ${\sigma }_Y^2$ in the denominator. These are population variances (${\sigma }^2=\frac{\mathrm{\Sigma }(x_i-\overline{x})}{n}$) and not sample variances ($S^2=\frac{\mathrm{\Sigma }(x_i-\overline{x})}{n-1}$). Using $S^2$ instead of ${\sigma }^2$ would lead to slight discrepancies.