In this paper, new approaches for chaotic time series prediction areintroduced. We first summarize and evaluate the existing local prediction models, then proposeoptimization models and new algorithms to modify proced...In this paper, new approaches for chaotic time series prediction areintroduced. We first summarize and evaluate the existing local prediction models, then proposeoptimization models and new algorithms to modify procedures of local approximations. Themodification to the choice of sample sets is given, and the zeroth-order approximation is improvedby a linear programming method. Four procedures of first-order approximation are compared, andcorresponding modified methods are given. Lastly, the idea of nonlinear feedback to raise predictingaccuracy is put forward. In the end, we discuss two important examples, i.e. Lorenz system andRoessler system, and the simulation experiments indicate that the modified algorithms are effective.展开更多
文摘In this paper, new approaches for chaotic time series prediction areintroduced. We first summarize and evaluate the existing local prediction models, then proposeoptimization models and new algorithms to modify procedures of local approximations. Themodification to the choice of sample sets is given, and the zeroth-order approximation is improvedby a linear programming method. Four procedures of first-order approximation are compared, andcorresponding modified methods are given. Lastly, the idea of nonlinear feedback to raise predictingaccuracy is put forward. In the end, we discuss two important examples, i.e. Lorenz system andRoessler system, and the simulation experiments indicate that the modified algorithms are effective.
