Abstract: In recent years, severe haze weathers have hit many areas of China, which has greatly affected people’s health and life. In the study on the haze problem, the relationship between PM_2.5 and other air pollutants has become the focus of academic researches. Therefore, based on the vector autoregressive model and in combination with unit root test, AR eigenvalue test, generalized impulse response function and variance decomposition method, the air quality related data including PM_2.5 in Tianjin from January 1, 2014 to December 31, 2016 were analyzed. Also the dynamic relationship between PM_2.5 and its influencing factors was analyzed and the impact of other atmospheric pollutants on PM_2.5 studied. The result shows the model of air quality system formed by PM_2.5 and other air pollutants is stable. The increase of SO₂, NO₂ and CO concentration will cause the increase of PM_2.5 concentration in a short time; the effect of SO₂ and NO₂ on PM_2.5 is more significant; the increase of O₃ concentration has an inhibitory effect on PM_2.5. Therefore, suggestions on adjusting the industrial structure and strengthening the governance of SO₂ were put forward as the primary solutions for the government of Tianjin to manage haze pollution.