Figure 2.

Simulation of Crooks’ fluctuation theorem. (A) Simulation with 1000 cycles. In black, the theoretical prediction; in red, the linear regression for the simulated data and, in green, the simulated points. Since the simulated data set adjusts pretty well to Crooks’ fluctuation theorem (4), Jarzynski’s equality (5) is fulfilled. (B) Simulation with 20 cycles and bootstrapping. The black line is the theoretical prediction (4) while the red line and shaded area are, respectively, the mean and the 99 % confidence interval of (4) after 1000 bootstraps of the driving error values obtained in a single run (which consists of 20 cycles).