针对股价预测中存在的不确定性、间断性、随机性和非线性等问题,提出一种TRSSA-ELM(Tent Random Walk Sparrow Optimization Algorithm-Extreme Learning Machine)股价预测模型。首先,采用自适应Tent混沌映射和随机游走策略对算法进行改...针对股价预测中存在的不确定性、间断性、随机性和非线性等问题,提出一种TRSSA-ELM(Tent Random Walk Sparrow Optimization Algorithm-Extreme Learning Machine)股价预测模型。首先,采用自适应Tent混沌映射和随机游走策略对算法进行改进,增强种群多样性和随机性,提高算法局部和全局的寻优能力。其次,使用单峰、多峰和固定维多峰测试函数对TRSSA(Tent Random Walk Sparrow Optimization Algorithm)性能进行了验证,相比于SSA(Sparrow Optimization Algorithm)、AO(Aquila Optimizer)、POA(Pelican Optimization Algorithm)和GWO(Grey Wolf Optimizer),TRSSA算法具有更好的收敛速度、精度和统计性质。最后,由于ELM(Extreme Learning Machine)模型随机生成权重和阈值,降低了预测精度和泛化能力,应用TRSSA算法优化ELM模型的权重和阈值,并用三安光电股票数据集对TRSSA-ELM模型进行了测试。实验结果表明,TRSSA-ELM模型相比于SSA-ELM、ELM、SVR(Support Vector Regression)和GBDT(Gradient Boosting Decision Tree),具有更好的预测精度和稳定性。展开更多
A random walk Metropolis-Hastings algorithm has been widely used in sampling the parameter of spatial interaction in spatial autoregressive model from a Bayesian point of view. In addition, as an alternative approach,...A random walk Metropolis-Hastings algorithm has been widely used in sampling the parameter of spatial interaction in spatial autoregressive model from a Bayesian point of view. In addition, as an alternative approach, the griddy Gibbs sampler is proposed by [1] and utilized by [2]. This paper proposes an acceptance-rejection Metropolis-Hastings algorithm as a third approach, and compares these three algorithms through Monte Carlo experiments. The experimental results show that the griddy Gibbs sampler is the most efficient algorithm among the algorithms whether the number of observations is small or not in terms of the computation time and the inefficiency factors. Moreover, it seems to work well when the size of grid is 100.展开更多
文摘A random walk Metropolis-Hastings algorithm has been widely used in sampling the parameter of spatial interaction in spatial autoregressive model from a Bayesian point of view. In addition, as an alternative approach, the griddy Gibbs sampler is proposed by [1] and utilized by [2]. This paper proposes an acceptance-rejection Metropolis-Hastings algorithm as a third approach, and compares these three algorithms through Monte Carlo experiments. The experimental results show that the griddy Gibbs sampler is the most efficient algorithm among the algorithms whether the number of observations is small or not in terms of the computation time and the inefficiency factors. Moreover, it seems to work well when the size of grid is 100.
