最后本文以厦华彩电近三年的销售数据为研究样本，对文中涉及的模型进行了实证分析。
These forecasting models in the paper are tested with the data on sales volume and market share of XIAHUA color TV in three years lately.
最后利用切削样本数据对网络进行了训练，并同分析模型在拟合精度与预测精度方面进行了比较。
Finally by training the network is trained using cutting dates, so the prediction precision of each model is compared.
最后取一组小样本数据进行计算，实例结果表明所提出的方法合理有效。
At last an example is given to demonstrate the availability and ap-plication of the proposed method.
最后以数值仿真得到的数据为样本数据，通过设计网络结构和选用学习算法，建立并得到基于BP人工神经网络的翘曲——收缩预测模型。
Finally, taking data from CAE as samples; the BP neural network of warping-shrinkage prediction model is established by designing the network structure and selection of learning algorithm.
先对采样数据进行分析处理，形成网络的样本数据库，然后对网络进行训练、仿真，最后再对仿真结果作进一步分析。
The sample data are firstly analyzed and processed to form a sample database of the network. Then, the network is trained and emulated. Finally, the emulation result is further analyzed.
最后，对研究思路与方法进行了说明，并说明了调查问卷的设计、样本的选取以及数据的处理等情况。
Thirdly, the methods being used in this research is displayed. Meanwhile, the author explains the designing of the questionnaire, the selecting of the samples and the dealing of the statistics.
同时将样本分为三个区间即2004年、2005年与2006年。最后对数据进行分阶段的统计特征分析、多因素回归分析。
Meanwhile divides them into three sectors namely 2004, 2005 and2006, then analyzing them the statistical characteristic and multiple regression analysis.
最后选取某些样本县，在样本县内对比每个乡镇图上的人口与实际统计人口，结果表明该人口分布数据有较高的精度。
Finally, the comparison of the statistical population density and the actual population density of the selected sample counties proves the higher precision of the results.
最后对训练样本数据个数、神经元个数的选择进行了探讨和经验总结。
Finally, discuss and research how to select the number of the neural units and the training data.
最后对训练样本数据个数、神经元个数的选择进行了探讨和经验总结。
Finally, discuss and research how to select the number of the neural units and the training data.
应用推荐