You are displaying the Monte Carlo means of the parameter estimates. Means are, of course, not robust statistics, so this could be caused by a small number of random samples for which one initial conditions (ICs) converges and the other does not (or converges to a different local maximum.)   Here's how to determine what is happening: Run the simulation with one set of ICs and save the parameter estimates to MLE_EST1. Then run the simulation again with the other ICs and save the estimates to MLE_EST1.  Plot the parameter estimates for the two runs against each other in a scatter plot.  For most of samples, the two estimates should fall on or near the line y=x.  For a few, you will probably see that the estimates are far from each other. If you label each marker with the simulation number, that will enable you to find out which samples are leading to different estimates. You can then investigate the MLE function for those samples. ... View more