Estimation Results           Rel err                                        exact measurements Θ1 0.501 ± 0.016 2.5% Θ2 0.00250 ±7·10-5 2.2% Θ2 0.301 ± 0.011 3.1%                                            noise: σ=10 Θ1 0.490 ± 0.019 3.2% Θ2 0.00248 ±9·10-5 2.9% Θ2 0.302 ± 0.012 3.4%                                            noise: σ=25 Θ1 0.454 ± 0.031 9.7% Θ2 0.00243 ±15·10-5 5.2% Θ2 0.301 ± 0.021 5.6%

Table 2: Statistics of the estimation results for Lotka-Volterra model. 50 data sets are simulated using the Gillespie algorithm with 40 observations with Δt=1 and true parameter Θ(0)=(0.5, 0.0025,0.3) and v0=(71,79). For each of the data sets an estimation is performed with the MSS method. The table shows a statistic of the 50 estimates: Parameter name (column 1), averages of estimates (column 2), standard deviation of estimates (column 3) and relative errors (column 4).