Research On Two-Stage Investment Decision-making Model Based On Scenario Simulation

At present, stage-investment has become an important and popular means of investment, which can greatly reduce the risk of investment. Because the investment is not reversible,the benefit of the investment is uncertain.while,the investor has many flexible choices when they face the investment problem. As a result, many investors often invest appropriate money in different stage of the project rather than invest all the money the project needed in the beginning when they invest in projects with long period and high risk.Stage-investment decision-making is to find a decision path which can make the value of the project maximum.however, the investment process is influnced by many uncertain factors, which result in the uncertain of the profit of the project. Besides, The decisions of different stage behave with the character of path-dependence : the decision maked in the pre-stage will affect the decisions that can be choosed in the subsequent stage;the superior decision to be choosed in the pre-stage depend on the effect of the decision choosed in the subsequent stage; while,the existance of the decisions in the subsequent stage depends on implementation of the pre-stage's decision . all of which makes the decision-making problem of stage-investment become very complicated .Now ,the means used to deal with decision-making problems of stage-investment mainly consist of dynamic programming,compound options pricing model,decision-tree,binomial tree,mento carlo simulation and so on .But theses methods can only be efficient in a limited range and can not reflect the decision-making process of stage-investment . we have to search an more applied method to resolve the decision-making problems of stage-investment . The dynamic programming can only solve the problems that the profit of the project can be forecast exactly ,besides, it hasn't a standard model and a universal method to construct model, even it hasn't a guide line to judged whether an investment problem can be dealed with the dynamic programming; The Geske compound option pricing model come into being under strict market assumption, but many investment problems are beyond the conditions, moreover, The Geske compound option pricing model can only solve the issues that the project is affected by only one uncertain factor; The decision tree analysis is only suitable for the project that the number of cash flow and its probability can be forecasted exactly;resembling the Geske compound option pricing model, the binomial tree is restricted by many strict condetions too;The Monte Carlo simulation can only solve the investment problems that in the whole process there is only one decision-making path , it cannot be used to deal with the project with the character of path-dependence. Therefore, we need to study a more practical and operational method to solve the stage-investment decision-making problem.We regard the project value as a dynamic accumulation process that is influenced by many uncertain factors. After rigorous mathematical reasoning,we find that the dynamic accumulation process of project value can be depicted as second-order stochastic parital differential equation set.While,the essence of systems dynamics is high-order stochastic parital differential equation set.In this paper,we take two-stage investment as our researchful object and go to look for the influence-factors of the project's value by analyzing the composing of the cash flows . and construct two-stage investment decition-making model using theory of systems dynamics , sequentially , we will simulate each decision-making path in the software of powersim studio for many times , and record the useful datas that can be helpful to find the optimal decisions ,which contain NPV of each stage for each decision-making path ; And then , we need compute the expection of each stage from these records ; finally ,we can find the optimal decision-making path in reverse order in accordance with bellman optimization theory . At the end of this article , we make use of this means to deal with a two-stage investmen project , which indicate that it works very well . the means that we discussed in this paper has many advantages including less restrictive conditions and more flexibility when dealing with the stage-investment projects that are affected by many uncertain factors .In this article, we analyse the characters of investment and the difficulties of decision-making in detail.we effectively deal with the problem of path-dependence in investment process and compounding many stochastic process, and eventually contrive the analytic frame that can be used to solve stage-investmet decision-making problem effectively, which is a new approch for stage-investmet decision-making problem .