Abstract: In this paper,according to the China pension fund situation,we develop the optimization dynamic investment strategy models based on the Bayesian stochastic programming approach,in which we improve the Bayesian vector autoregressive by using the Minnesota Prior.According to the improved model,we estimate the asset future returns and give the concreted calculation steps for solving the models.Finally and combining with the historical data,we conduct a simulation,the result shows that the optimal investment strategy can be solved according to the reality.

Key words: stochastic programming, investment strategy, Bayesian vector autoregressive