The application of synchronous phasor measurement technology provides an important technical basis for distribution network estimation such as power flow Jacobian matrix estimation and voltage/phase angle-power sensitivity estimation. Considering the correlation, sparseness and symmetry of the power flow Jacobian matrix, a sparse estimation method of power flow Jacobian matrix and sensitivity matrix based on the measured data of a synchronous phasor measurement unit is proposed. It can effectively estimate the Jacobian matrix and the sensitivity matrix with less measurement. To tackle the problem of bad data appearing in the measurement process, a more robust weighted least squares method is introduced to improve the robustness of the algorithm. Finally, the feasibility of the method is verified on an IEEE33 node power distribution system. This work is supported by National Natural Science Foundation of China (No. 51937005).