The best-fitting lines for these three plots were calculated with vertical residuals (green lines in Figure 2).
Linear mixed models deal with missing data by predicting the best-fitting line for each patient without data imputation.
We computed the slope of the best-fitting line for each participant and then determined whether all slopes were different from zero.
Symbol (O) represents the experimental data and solid line is the best fitting line to Eq. (1).
For all cases, Fig. 1 shows that the power spectra follow a power-law form typical of a scaling process, with the global exponent β corresponding to the slope of the best fitting line in the log-log plot.
What to do? Plot them on a graph and calculate the best-fitting line, of course.
Using mean plots and one way analysis of variance tests, we observed that the best fitting trend lines for these associations were linear.
In 5 of the 6 subjects, the best-fitting line accounted for 95% or more of the variance of the data (R2 values ≥0.95); in the sixth subject it accounted for 89% of the variance.
Different scholars choose different estimates, with the result that their best-fitting lines differ.
The best-fitting lines to our two sets of four time shift estimates are shown in Figure 4c.
