Considering that the variation of nonlinear characteristics of DC-DC converter with parameter k and the fact that entropy can reflect the statistical characteristic of numerical sequence, in this paper we propose a new method to estimate the nonlinear characteristics of DC-DC converter based on the information about entropy. In this study, taking one-order DCM Buck and Boost converter for example, the distributions of numerical sequences and entropies are analysed under different values of feedback gain k and initial values of x0. The results prove that the entropy of DC-DC converter is determined by feedback gain k and irrelevant to initial value x0, and its final value is smaller than the theoretical maximum value log2N (N is the number of partition), and that entropy can discriminate the period-doubling bifurcation and chaos. So a quantifiable nonlinear dynamical behavior index is obtained which can provide theoretical basis for understanding chaos characteristic and chaos control, and perfect the theoretical analysis method in DC-DC converters.