The theory analysis and numerical experiment indicate that the combined metamodel approach has some good characteristics such as high fitting precision , good prediction ability , insensitiveness for simulation data , etc 理论分析和数值实验表明，组合元模型方法具有拟合精度高，预测能力强，对仿真数据敏感性弱的特点。
Uniform design method is used to constitute test examples . back test is used to test the result . the result proves that forecast capability of neural networks is good enough 神经网络训练结束后，采用均匀试验设计方法构造测试样本对神经网络进行测试，并采用后验差检验方法进行结果检验，证明神经网络预测能力较好。
This approach uses word prediction capabilities and introduces a concept we call speech - enabled word prediction . in effect , spoken words are displayed as a list of choices enhanced by word predictions 这种处理方式具备了文字预测能力，即引入了我们称之为"语音化单词预测"的概念，最终效果是，听写功能通过单词预测列表显示得到了加强。
By means of such default ratio model , not only default companies and credit companies can be recognized successfully , but also default action can be predicted one and a half years in advance 结果表明，基于期权定价理论的违约率模型不但能够有力的识别违约公司和非违约公司，而且可以提前1 . 5年反映公司的违约状况，具有较好的预测能力。
It is proved that the semi - analytical prediction method makes the model has a better prediction precision , a better generalization ability and a lower dependence on the number of samples by using matlab tool 通过仿真分析,证实了应用半解析方法进行预测时,模型的预测精度更高,泛化性能更强,对样本的依赖程度更低,并具有更好的长期预测能力。
Ability to interact and manage people ; scheduling and forecasting abilities ; knowledge of statistical process control and other tqm techniques ; selfmotivated ; understanding of role in the overall corporate goals 能影响及管理人员；有安排及预测能力；有统计过程控制及其他全面质量管理技术方面的知识；自身有高度积极性；能明确自身在公司目标下的角色。
We find that current asset turnover , debt ratio , revenue growth rate , profit margin before taxes and investing gains , return before taxes on assets et al . can predict financial distress accurately 研究发现,由流动资产周转率、资产负债率、销售收入增长率、扣除投资收益的税前销售利润率和税前资产利润率这五个财务指标建立的数学模型具有较强的预测能力。
The basic characteristics of chaos are presented first in this paper , then the rbf neural network is trained using chaos time series and afterwards the trained neural network can be utilized to predict the chaos series of future time 摘要介绍了混沌的基本特性，利用混沌时间序列对rbf神经网络进行训练，则训练好的神经网络对混沌序列未来时刻的值具有一定的预测能力。
Observing the total error curve , the convergent speed is faster , and the error precision is higher then a group untrained experimental data are input the simulative system in turn . and their predicting results are satisfying 根据小型催化裂化装置进行重油催化裂化反应试验的试验数据对系统进行预测能力分析，从全局误差曲线图可见训练速度比较快，误差精度比较高。
( 5 ) state anxiety is affected strongly by " self - blaming " . trait anxiety is influenced strongly by " self - blaming " , " solving problem " and " imaging " . the students with higher anxiety tend to use negative coping style ( 5 )自责对状态和特质焦虑都有较强的预测能力，自责、解决问题和幻想因子对特质焦虑的预测能力较强；高焦虑的大学生倾向于运用消极的应付方式。
