In doing the above experiments, it was observed that the relative error in the variance of the final position as \(n\) increased does approach the true variance \(2Dt\) but the convergence is not smooth. As the relative error (around \(5\%\) to \(10\%\)), then increasing \(n\) more can cause the error to sometimes increase and not decrease as one would expect. Meaning the relative error is not monotonic decreasing as \(n\) increases. However, as \(n\) becomes very large, the trend is for the relative error is to decrease. I can only contribute this behavior to some sort of statistical error. This needs to be investigated more.