Efficient portfolio management, has been attractive for financial researchers and was wished for investors from past to now. In this research, a multiperiod portfolio optimization problem for asset liability management of an investor who intends to control the probability of bankrupt is investigated. The proposed portfolio is consisting of number of risky assets, risk free asset and a type of debt. A mean variance model, with constraint of bankrupt controlling in different time horizons is proposed. Lagrangian Multiplier Method with dynamic programming is used for solving proposed model and regarding to its complexity degree, Genetic algorithm was the best selection for reaching numerical results. The proposed model is ran with real data consisting of 10 accepted company in Tehran stock exchange, bonds and bank loan as an investor debt.
