三项型指数曲线在“药-时”曲线拟合中的应用  

The Application of Trinomial Exponential Curve in the Fitting of Drug Concentration-time Curve

作　　者：高辉[1,2] 胡良平[1] 鲍晓蕾[1] 

机构地区：[1]军事医学科学院生物医学统计学咨询中心 [2]解放军95969部队卫生队

出　　处：《中国卫生统计》2012年第5期677-679,共3页Chinese Journal of Health Statistics

摘　　要：目的探讨三项型指数曲线在"药-时"曲线拟合中的应用及其SAS实现。方法借助SAS软件,调用REG过程以常规的残数法得到曲线上各参数的估计值,然后以这些估计值为初始值,调用NLIN过程进行非线性最小二乘估计,得到优化的曲线回归方程。结果残数法与非线性最小二乘法相结合所得最终曲线回归方程较仅用残数法计算而得的曲线回归方程有更好的拟合效果。结论运用高级SAS编程技术可便捷地对"药-时"关系资料实现三项型指数曲线的拟合,且拟合效果较好,具有广阔的应用前景。Objective To explore the application of trinomial exponential curve in the fitting of drug concentration-time curve and its SAS realization. Methods Applying the traditional residual method to the data by invoking the REG procedure in SAS to get the estimated values of the curve parameters, then perform the method of non-linear least squares by invoking the NLIN procedure to the initial estimated values to optimize the curvilinear regression equation. Results Compared to the residual method, the combination of residual method and the method of non-linear least squares fits a better final curvilinear regression equation. Conclusion SAS program can easily realize the trinomial exponential curve in the fit- ting of drug concentration-time data. It fits the model well and has a wide application prospect.

关 键 词：三项型指数曲线 “药-时”曲线 曲线拟合残数法非线性最小二乘法 

分 类 号：R195[医药卫生—卫生统计学]