Atmospheric disturbance influence the measurement accuracy greatly in laser precision measurement of long distance. A real-time filtering algorithm based on time series analysis theory is proposed in this paper. Firstly, ARMA (Auto-Regressive and Moving Average) model of sample data is set up based on the theory of time series analysis, then the mathematical expectation is obtained according to the model as the filtering result, finally, it is integrated into the laboratory laser beam drift measurement system for real-time filtering measurement. It is evident from the simulation results that high accuracy can be obtained after filtering. The experimental results show that the stability of the filtered drift increases by about 20%.