The operation replaces all the missing values from the source data by using the values found from the nearest data record. The nearest data record is determined by using the normalized distance that is computed according the following formula:
where is the number of such components for which values are missing in r(a) or in r(b), and N is the number of data fields. The computation of the distance notices the missing values, and it scales the distances such that the pairs of the data records including more missing values get larger distances.
The operation msdbycen replaces missing values by the values of the prototypes given in a separated data frame. The classified data must tell which data records are connected to which prototypes. See also the command somcl 5.1.2 with the missing data flag.