This paper studies the mechanism of mutual evolution in the stock market,money market and foreign exchange market in China.Based on monthly data from January 2000 to May 2014 in Chinese financial markets,the cause-effect relationships between 19 important indexes are tested,and 12 decision indexes which not only can reveal their market conditions but also are closely related to the other two markets are obtained.Then,this paper designs three composite indexes for the three markets respectively and proposes two nonlinear evolution models based on ordinary differential equation for the three markets.Furthermore,a COMDE algorithm is given to solve these models.The empirical results show that the three composite indexes can better reflect the fluctuations of the markets,and the constrained non-linear evolution model can better depict the non-linear evolution structure among the three markets and the interaction relationships.The results can provide decision supports for govemments and investors.

Aiming at the problems that neural networks are easy to fall into over learning and the number of nodes are difficult to determine,we establish a double-layer DE-ELMmodel.This model sets the number of nodes and the hidden layer node parameters of ELM as the individuals in outer and inner layer of DE algorithm,and then uses natural selection mechanism to select the best parameters.We make a short-term and medium-term prediction with Shanghai Composite Index and S&P 500 Index by double-layer DE-ELM prediction model,and compare with DE-ELM,ELM and BP neural network model.The empirical results show that this prediction model can select the pa-rameters of ELM model more effectively,and it has better prediction ability and better stability than comparison models.

This paper proposes a bondage-based structural model that is derived from the trilemma constraint among the monetary,foreign exchange and securities markets.It introduces a foreign exchange intervention index to explore a new market landscape and regulatory strategies.Based on the differential evolution algorithm and its application to the data from January 1998 to April 2014,it presents the estimated nonlinear systems among the three policy goals of monetary independence,foreign exchange intervention and capital account openness.The analysis results show the existence of equilibrium and the local structural stability of systems by the system dynamics approach.Capital account openness and monetary independence are in proximity to moderate levels after a short-term evolution (about 1 year).It could lock-in into different proportions in the long-term evolution (about 5 years) for its sensitive dependence on initial conditions.In addition,China has the policy preference in the predominance of exchange rate stability while the capital openness could be the leading indicator to affect their proportions.

基于“三元悖论”提出货币、外汇和证券三个主要市场间的“束缚”结构模型，引入外汇干预指数，探讨在外汇储备政策导向下市场的新格局及监管策略．以1998年1月至2014年4月的数据为驱动，采用差分进化算法估计货币政策独立性、外汇干预指数与资本开放度三个政策目标的非线性系统．利用系统动力学方法分析得到：系统存在平衡点并具有局部结构稳定性，资本开放度与货币政策独立性的指标在短期演化后（约1年）趋于中等水平；系统同时表现出初值敏感依赖性，具有不同初始值的各指标经长期演化后（约5年）趋于不同的配比。另外，我国在政策偏好上重视汇率的稳定性，而资本开放度可作为影响三个指标间配比的前瞻性指标。

The basic concept of Analytical Hierarchy Process(AHP) is introduced,and the computation methods to AHP and its applications are analyzed in this paper.As to AHP, there are the following four computation methods:Geometric Average(GA),Arithmetic Average(AA),Eigenvector Method(EM),and Least Squares(LS).In most literatures,one of the four computation methods of AHP is used to solve weight vector in analyzing some practical problems.As a result,different methods can induce different results with warps.In this paper,four computation methods are used to solve weight vectors of a practical example, respectively,and then these corresponding results are sorted order and synthetically analyzed in order to avoid the warps produced by using one method.Furthermore,the comprehensive and effective results are obtained.