Figure 3.

Errors of the mean and standard deviation of the total substrate x_t vs k₃ for the different approximations discussed here. Except for k₃, we used the same parameters as in Fig. 2. We used 2000 Monte Carlo realizations for the smallest value of k₃ and increased the number of realizations linearly, working with 20 000 realizations for the largest value of k₃ shown; this was done to maximize accuracy for the available computational power. Our method is uniformly more accurate as k₃ is varied.