The size of the cohe_ci and phase_ci params for mt_coherence function?

[jiezhou87]

Lately I tried to use the mtspec to calculate the coherence between two time series. And I also want to estimate the uncertaity of the estimation. Based on Prieto's paper and the code, the mt_coherence should return cohe_ci and phase_ci with size (nf,2). However what I get is only size of nf. Did I used it some places incorrect?