The authors investigate cobweb type service price fluctuations and adaptive learning, and put forward a dynamic service pricing method based on the dynamic stochastic wealth model (DSWM). In this model, authors treat the service market as an expectation feedback system, where the current market equilibrium price of service determined by aggregated individual service providerpsilas expectations of next periodpsilas yield. After several rounds of price adjustment, optimal pricing strategies can be conducted. In order to describe the fluctuations of price in future, authors also developed a set of market impact functions with the technique of dynamic harmonic regression modification to simulate the market impact.