Abstract: The effects of macro-economic factors on stock markets are analyzed using Markov Switching Vector Autoregression model. The factors of output, price and monetary are chosen to reflect economic growth and monetary policy, which include industrial added value, consumption, export, inflation rate, M2, interest rate and real exchange rate of RMB against US dollar. Firstly, the states of factors in different periods are identified by Markov Switching Vector Autoregression to investigate the relation of macro-economic factors and stock price separately. Further, by building Markov Switching Vector Autoregression for absolute difference to get impulsion response function of macro factors fluctuation to stock market, the effects of different macro-economic factors on stock markets are compared. Finally, the conclusion and policy suggestions are got from empirical results.The conclusion demonstrates that the return of China's stock market has obvious nonlinear relationship with the macroeconomic variables,The stock market reflects the overall health of the economy; in most of the time the real economy changes and stock market volatility has maintained good cooperativity, the stock market cycle was slightly ahead of the real economy cycle; the rapid growth of M2 was often accompanied by a sharp rise in the stock market, the frequent changes in monetary policy have a significant effect on the stock market; interest rate adjustment did not have an immediate impact on the stock market cycles, there exists an inverse relationship between interest rates and stock price; greater volatility of stock movements was correlated with higher inflation rate; The periodicity of the real exchange rate changes is consistent with the stock return cycle.

Key words: Markov switching vector autoregression model, economic cycle, M2, interest rate, exchange rate