[gmx-users] How does g_msd calculate the deviation of diffusion coefficient?

[chris.neale at utoronto.ca]

> I know that g_msd calculates the D using the least square (LSQ) method.
> But I dont know how to calculate the deviation of D. In this case, how
> does g_msd get +/- 0.0366?

This snippet can be found by g_msd -h or on the online documentation:

An error estimate given, which is the difference of the diffusion  
coefficients obtained from fits over the two halfs of the fit interval.