| 基于神经网络的股市资产收益率识别 |
| |
| 引用本文： | 林胜乐,陆杨.基于神经网络的股市资产收益率识别[J].价值工程,2005,24(10):119-123. |
| |
| 作者姓名： | 林胜乐 陆杨 |
| |
| 作者单位： | 大连理工大学管理学院,大连,116024;大连理工大学管理学院,大连,116024 |
| |
| 摘 要： | 本文以BP神经网络来识别中国股市不同行业未来的资产收益率,来检验财务数据的有效信号的假定,输入神经元选取了安全性指标、盈利性指标、成长性指标、现金流量结构分析指标四大类指标,共15个最为经典的财务指标,用以预测不同行业的未来超额收益率。预测结果显示,平均预测误差为0.7%,表明神经网络在分析行业未来盈利能力有很强的分类功能。模型分析证实财务数据在行业层面上有十分强烈的信号。同时,预测方法为投资者掌握未来投资资金流向有实用价值。
|
| 关 键 词： | 神经网络 资产收益率 财务指标 |
| 文章编号： | 1006-4311（2005）09-0119-05 |
BP Neural Network Recognition and Proof-test of Accounting Signals in Chinese Stock Market |
| |
| Lin Shengle,Lu Yang.BP Neural Network Recognition and Proof-test of Accounting Signals in Chinese Stock Market[J].Value Engineering,2005,24(10):119-123. |
| |
| Authors: | Lin Shengle Lu Yang |
| |
| Institution: | Department of Economics, Dalian University of Technology, Dalian 116024, China |
| |
| Abstract: | This work aims at proof-testing the effectiveness of financial statements at Chinese Stock market, via forecasting the future market-adjusted abnormal returns for different industries on the basis of previous year's financial statement ratios.15 ratios are selected following four categories of indicators: stability, profitability, growth potential and cash-flow distribution. The application of BP neural network in forecasting produces high precision, with a mean deviation of 0.7%(absolute value) and extreme small deviations for the most and least profitable industries. This paper demonstrates that the financial statements are informative at the industry level. Besides, neural network can be a reliable tool as forecast for investment. |
| |
| Keywords: | BP neural network asset return financial ratios |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
