Redshift dependence of GRBs' observed parameters

GRBs are extremely energetic short cosmic transients. Due to their huge energy output in a short time they can be observed at very large cosmological distances. Actually, they sample the whole observable Universe. As a consequence of their large distances, their observed duration, fluence and peak flux depend on the redshift. In the reality, however, this dependence can be observed only in the case if the intrinsic variance of these quantities in comov-ing frame do not exceed significantly that coming from different redhifts of the GRBs. Nevertheless, it is an important question whether the redshift de-pendence of the observed quantities could be extracted from the observational data. Using a training set consisting of GRBs having measured physical param-eters and redshifts we are looking for the effect of the redshift on the observed data, using techniques available in multivariate data analysis. Creating a 3D parameter space from duration, fluence and peak flux, we define partitions in the distribution of data points and compare the redshift distributions within these partitions. Partitioning will be made by some hierarchical clustering algorithm and cutting the obtained agglomeration tree at different places to get partitions of different numbers. The distributions of redshifts within the partitions, obtained in this way, will be compared to see if there is any difference in redshift distribution between partitions at all.