摘要
讨论了非线性模型ｙ＝ａｘｂ的对数线性化回归方程参数的普通最小二乘估计．指出估计无效的原因是对数线性模型的随机误差具有异方差性质．运用二步加权最小二乘法改进了参数的估计．并在模拟试验中证明了其优越性．
In this paper,we discuss the ordinary least square estimate of the parameter in the logarithmetics regression equation of non-linear model y=axb.We point out the reason why the estimate is no avail is that the random error of the logarithmetics regression equation has dif ferent variance character.In this paper,we improve the estimate by Two Stage WLS Method. It is proved in the imitate experiment that the improved estimate is be tter than the ordinary least square estimate.
出处
《湘潭大学自然科学学报》
CAS
CSCD
2001年第3期1-5,共5页
Natural Science Journal of Xiangtan University
关键词
普通最小二乘法
异方差
加权最小二乘法
二步加权最小二乘法
Ordinary least square method
Different Variance
We ighed least squared methed
Two stage WLS method. 1991 Mathematics Subject Classification:Primary 62J02
