A data processing apparatus, including at least: a register for storing a data set W; a processor, coupled with the register to divide the data set W into a plurality of groups according to an experimental independent variable set V, |V|≥1; use a dividing method yx, where (x,y) belongs to a dividing parameter set L={(x,y)| both x and y are positive real numbers}, |L|≥1, to divide each of the plurality of groups into a plurality of regions in a space of a dimension determined by an element of a data variable set Q, |Q|≥1; perform a statistical analysis with respect to an element Du of a dependent variable set D on the plurality of regions of each of the groups, where Du∈D and |D|≥1; and output a statistical result set R.