Submitted by yubing on Fri, 03/17/2017 - 16:19
各位专家,大家好
我在做fMRI相关分析的时候,有文章提到说,计算两个时间序列相关性的时候,需要考虑到时间序列内部自相关(autocorrelation的影响
文章内容如下
"Because individual time points in the BOLD signal are not statistically independent, the degrees of freedom must be corrected according to Bartlett’s theory"(Michael D. Fox,PNAS,2005)
“Pearson correlation maps for each cortical ROI and sleep stage were converted to Z-maps using Fisher's transform with the degrees of
freedom corrected using the Bartlett correction factor (BCF; Fox et al.,2005; Zhang et al., 2008;Hale et al., 2015) to account for autocorrelation
(BCF = 1.65 ± 0.03, mean ± standard error across participants).(Joanne R. Hale,Neuroimage,2016)
不知道REST的 detrend 是否消除了 autocorrelation的影响?多谢指教!
Submitted by ZangYF on Sun, 03/19/2017 - 20:15 Permalink
Re: 相关分析的时候,自相关影响的去除
去除自相关,只是众多协变量一种。就像去除全脑平均时间序列一样,去与不去,各有各的道理,并没有一个金标准说明哪一个更好。
Submitted by yubing on Mon, 03/20/2017 - 08:21 Permalink
Re: 相关分析的时候,自相关影响的去除
多谢臧老师指教!