摘要
目的 :研究医学重复观测数据的多变量随机系数模型 .方法 :对两种药物 (A药 :消瘾扶正胶囊 ,B药 :可乐宁 )治疗 12 0例患者后的舒张压和收缩压重复观测数据进行多变量随机系数模型分析 ,对模型系数的固定效应参数矩阵 ξ作最小二乘估计并进行组间比较 ,同时估计随机效应的方差 协方差矩阵 ,分析方法用SAS/IML软件编程得以实现 .结果 :得到了固定效应和随机效应有关参数的估计值 ,并给出了曲线图 .用药后患者的舒张压和收缩压随时间的变化而变化 ,且两个药物组曲线的变化趋势是不相同的 ,A药组的变化相对平缓 ,而B药组起伏波动较大 ,用药后A药组的舒张压和收缩压相对来说均较B药组为高 .结论 :多变量随机系数模型可有效地进行多变量重复观测数据的动态变化趋势分析以及随机效应分析 .
AIM: To study multivariate random coefficients model of repeated measures data in medical research. METHODS: Both diastolic and systolic blood pressures repeated measures data, collected from 120 drug abusers after taking two kinds of medicine (Drug A: Xiaoyinfuzheng, Drug B: Kelening), were analyzed by multivariate random coefficients model. The fixed effect parameters matrix x of model coefficients were estimated by using least squares estimation method, the effects between treatment groups were compared and the variance-covariance matrices of random effect were also estimated. Related analysis methods were programmed with SAS/IML code. RESULTS: Estimated parameters with fixed effect and random effect were obtained and graphs were drawn. Both diastolic and systolic blood pressures changed with time after treatment and the trends between treatment groups were different. A slow change was observed in Drug A group, while a greater curvature was found in Drug B group. Both diastolic and systolic blood pressures in Drug A group were higher than those in Drug B group. CONCLUSION: Multivariate random coefficients model can effectively analyze the dynamic change trend and random effects of multivariate repeated measures data in medical research.
出处
《第四军医大学学报》
北大核心
2004年第23期2182-2185,共4页
Journal of the Fourth Military Medical University
基金
国家自然科学基金 (3990 0 1 2 6)
陕西省自然科学基金(2 0 0 3F1 1 )
关键词
重复观测
随机系数模型
多元统计学
repeated measures
random coefficients model
multivariate statistics
