遇到个问题亟需解决,可是本人学工科出身,尤其当年概率统计学的不好,所以想在这求各位帮忙看看,给点意见,先谢过看到本帖的各位了。
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
问题是这样的,需要比较两条曲线的差异是否显著?
现在有很多文献采用一种方法是比较曲线下的面积(Area Under Curves, AUC)是否有显著差异,如图1.jpg所示的这种情况,显著性比较敏感(经常得到p<0.01),因此比较容易就得到两条曲线有显著差异。
此主题相关图片如下:
但是如果是图2.jpg的这种情况,两条曲线出现交叉重叠时,采用AUC方法也能得到p<0.05的显著差异,但是实际情况很难判断出显著差异性的。
此主题相关图片如下:
我在一篇最新的文献上找到一模一样的问题的解决方法,如下所述:
Growth curve modeling, with an intercept (baseline), linear and
quadratic terms, was used to compare the ATest values from 0.1 to 0.6 in the two subject
groups. This analysis was performed using the Mplus software (http://www.statmodel.com).
Growth curve models describe change (growth) with respect
to a control variable. They are well-suited for analyzing grouplevel
differences in biomedical data, particularly in cases where
capturing and analyzing individual growth trajectories is important.
In our study, the growth trajectories of clustering coefficient
of a subject carry important information about the variance within
the group and needs to be incorporated in the model. The
coefficients of growth curve models capture the baseline
performance, instantaneous growth rate, and the acceleration of
the variable of interest–c.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
我想求助两个问题:
1,我在论坛上搜索了一下有关Growth Curve Modeling的帖子,发现大多数是讨论怎么样把一条曲线的数据拟合成“成长曲线”(Growth Curve),因此关于怎样采用Growth Curve Modeling比较两条曲线的差异性并没有太多的讨论。
我想知道的是,上面英文所述的方法中,怎么样具体实现比较“两条曲线的差异性”?比如:先将两条曲线各自回归成成长曲线,再具体比较什么参数能够得到显著性的概率?
2,上面提到的Mplus,我在论坛上也看了看,用的人比较少,也很难下载的到。
SPSS最新版(V17)是否有这种功能?我对统计的软件只会用最简单的SPSS,其他SAS、Mplus要编程才能实现的不太会用,如果SPSS有这种功能的话,具体怎么操作。实在对统计的东西有点白痴,各位xdjm多多帮忙,在此先谢过了!
本文来自: 人大经济论坛(http://www.pinggu.org) 详细出处参考:https://bbs.pinggu.org/thread-461422-1-1.html