-
mean = -7.651 +/- 0.028
-
A cutoff of 200 would be a good cutoff.
There appears to be some long correlation on the order of several hunderd
steps. If possible, one would want to run this longer to make sure the
effects of the initial transient were gone, and that phase space had been
sufficiently explored.
-
mean w/cutoff=200 = -7.630 +/- 0.020
Note also the reduction in the error.
With the errors, we can decided whether or not the difference between
these two averages are significant. In this case, -7.630 - (-7.651) = .021
is less than sqrt(.0202 + 0.0282) = 0.034, and the
difference is not statistically significant.error = sqrt( (error(a)^2 +
error(b)^2)/4 ) = 0.047