Third, the full model F statistic is computed using all of the calculated IRFs, the input stimulus functions, and estimates of baseline, trend, and noise to relate how well the data fits the full model.
Two regression models are fit: the baseline model (polynomial plus any -stim_base regressors), and the full model (baseline plus regressors including the IRF). The full F statistic is a measurement of how much the full model is an improvement over the baseline model. If the baseline model fits very well, then the full F might be small, since the IRF might add little explanatory power.
bob cox