标题: 用SPSS进行多分类Logistic回归分析时出现警告!帮忙看看吧! [打印本页] 作者: jingdian8893 时间: 2012-2-26 10:32 标题: 用SPSS进行多分类Logistic回归分析时出现警告!帮忙看看吧! 我用SPSS做微浸润癌、浸润性癌与原位癌的超声对比,病例数为微浸润癌(37例)、浸润性癌(83例)与原位癌(39例),观察指标为年龄(0=<40岁,1=40岁~,2=50岁~ ,3=60岁~)、肿块大小(0=<2cm, 1=2cm~, 2=3cm~)、纵横比(0=<1,1=≥1)、形状(0=规则,1=不规则)、边界(0=清晰 ,1=不清晰)、毛刺/蟹足(0=无,1=有)、恶性环(0=无, 1=有)、内部回声(0=不均匀,1=均匀)、钙化灶(0=无,1=有)、肿块后方回声(0=无变化,1=衰减,2=增强)、血流分级(0=0和1级,1=2和3级)、 LN转移(0无转移,1有转移),共12个观察指标,采用多分类Logistic回归分析,结果出现如下警告:
There is possibly a quasi-complete separation in the data. Either the maximum likelihood estimates do not exist or some parameter estimates are infinite.
The NOMREG procedure continues despite the above warning(s). Subsequent results shown are based on the last iteration. Validity of the model fit is uncertain.