$\mathbf{r}_i(t) \simeq \mathbf{x}^{(diffusion)}(t) + \mathbf{v}_{com} t$

$av(k) = \frac{1}{n} \sum_i P_i^{(run k)}(t)$

$\sigma(av(k)) \simeq \frac{1}{\sqrt{100}} \sigma(P_i)$

---

$\sqrt{\frac{1}{N}\sum_i \left(\frac{\hat{y}_i - y_i}{\sigma_y}\right)^2}$

$\frac{1}{\sigma_y}\sqrt{\frac{1}{N}\sum_i \left(\hat{y}_i - y_i\right)^2}$

$\frac{VAR_\mathrm{err}}{VAR_\mathrm{tot}}$

$R^2$?