Linear Regression
The connectivity spread measure was modelled to linearly vary with age and change in power. This was done separately for each frequency band and seed electrode group. Robust linear regression was used to minimize the effect of outliers (“Regression,” 2009), using the ‘fitlm’ function in MATLAB.
\begin{equation} \text{Conn}\left(f\right)=b_{c}+b_{\text{age}}*age+b_{\text{power}}*power\left(f\right)\nonumber \\ \end{equation}
The p-value for each of the coefficients was computed for the t-statistic of the two-sided hypothesis test \((H:b=0)\), given the other terms in the regression model.