Test of the Hypothesis That One Group of Dependences is Consistent with Another Group of Dependences

In this paper we describe the basic algorithmic characteristics of the test of the hypothesis that one group of dependences is consistent with another group of dependences for a case when the error disturbances of this data have normal distribution. Unpaired and paired versions are presented. The aim of our work was to find effective algorithms for biomarker identification in mass (MS) or Raman spectra (RS). This approach has been applied to mass spectral data and measured for identification of kidney carcinoma biomarker. Further more, the algorithm has been applied to Raman spectral data with the aim of recognizing the efficiency of heat denaturation of calf thymus DNA. The results of the applications show the need to find new regression algorithms, for example in non-parametric statistics.