| 固定影响平行数据模型参数的偏最小二乘估计 |
| |
| 引用本文： | 李卫国,王小伟. 固定影响平行数据模型参数的偏最小二乘估计[J]. 科技信息, 2008, 0(24) |
| |
| 作者姓名： | 李卫国 王小伟 |
| |
| 作者单位： | 北京航空航天大学理学院数字信息与行为教育部重点实验室 |
| |
| 摘 要： | 当平行数据的解释变量之间存在严重的多重共线性,或者样本点个数与解释变量个数相比较少时,用经典方法求解模型(如最小二乘法等)误差偏大,难以满足实际要求。针对这种情况本文提出用偏最小二乘法思想求解固定影响平行数据模型,并且实例表明误差明显减少,可以满足建模、预报的要求。
|
| 关 键 词： | 平行数据 偏最小二乘 多重相关性 |
PLS estimation and application of the fixed effect of panel data model's parameters |
| |
| Li Weiguo Wang Xiaowei. PLS estimation and application of the fixed effect of panel data model's parameters[J]. Science, 2008, 0(24) |
| |
| Authors: | Li Weiguo Wang Xiaowei |
| |
| Affiliation: | Li Weiguo Wang Xiaowei |
| |
| Abstract: | when the explanatory variables of panel data exist serious multicollinearity ,or the number of Samples is very small compared with the number of explanatory variables, classical methods(such as method of least minimum square) can’t be used. Otherwise the error is very large,which can’t satisfy actual request. Because of this complexion, this article brings forward PLS estimation of the model of fixed effect of panel data, which makes the error decreases. And it can satisfy the demand of modeling、prediction. |
| |
| Keywords: | panel data PLS multicollinearity |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
