Results are presented for four methods and three datasets: method A is a combination of Beagle 2.1.3 and DAGPHASE with scale and shift parameters equal to 2.0 and 0.1; method B is the same as method A, but with scale and shift parameters equal to 1.0 and 0.0; method C is Beagle version 3.3.0; method D is DAGPHASE using DAG from method C.
Computation time was highest for method B, DAGPHASE/Beagle with scale and shift parameters equal to 1.0 and 0.0, respectively.
In dataset 1, with 488 reference animals genotyped with the BovineHD chip, Beagle and DAGPHASE with scale and shift parameters equal to 2.0 and 0.1 (method A) resulted in an average imputation error rate of 1.91% for the six analyzed chromosomes.
Implicit parameters are method parameters which do not have to be explicitly passed to the method when it is called.
The sum of these three parameters equals one.
After the flattening procedure described in Section 3.2, we apply the method "alphashape" [ 45] with a parameter equal to 5000 for construction of a nonconvex hull.
Then, the taper parameter equals 0.05 in MZT method and CMT-MP-QP method to obtain approximate beam pattern null width of 8°.
The number of free parameters in the eigenvoice method is equal to the dimension (K) of the speaker subspace, which is much fewer than that of the MAP and MLLR methods.
In a special parameter setting, the λ parameter equals the reliability index, thus allowing convenient implementation of the method.
In addition, note that the saturation root mean square, average error and standard deviation decrease as we go as the following methods are used: (1) conventional methods to (2) CAPE method with parameter (a) equals to one, (3) 3D method, and (4) CAPE with, a, variable.
