Abstrakt
There is no doubt at present that computerized technologies in medicine and biological research, e.g., proteomics and genomics, need new approaches. This paper deals with Using Markers to Aid Decision Making in Diagnostics when data error disturbances have normal distribution.
The principles discovered can generally be useful in analogical dependence studies. The proposed algorithm is applicable generally to arbitrary problems from branches of other kinds of human activity formally of the type Using Markers to Aid Decision Making in Diagnostics.
It is necessary to view the proposed algorithm as a primary methodology pattern. The reason for this is that one must consider the precondition of data error disturbances having normal distribution.
Using the proposed algorithm, then, can be limited. Deriving the appropriate robust-mathematical-statistical analogues in meaning (e.g.) of R-estimators is a way out of this methodological difficulty.
Particularly in R-estimates, algorithms are not based on directly concrete values but on their rank statistics. In such cases, therefore, the requirement of normality in data error disturbances fails.