Autoregressive model and Nonparametric PAVA

When using the autoregressive model of general order it was found to require orders of up to 6; figure 12 shows the histogram of different AR orders required, pooled over all six of the null datasets taken with TR=3 secs. The results of using the autoregressive model and the nonparametric PAVA with the TR=3 secs data are shown in figure 15, along with Tukey with $M=10$ and multitapering with $NW=13$. Particularly for the boxcar design the single taper Tukey with $M=10$ performs the best.

Figure 12: Histogram showing the orders of the AR models required to estimate the autocorrelation for all voxels in all six null datasets (TR=3 secs) after applying the randomized ISI single-event design