Dear all, just some remarks on the topic: -Maximum Likelihood, in general, is not unbiased. But we should use an unbiased method at least for our cps (of course, nonlinear equations are unbiased in general) -expectation of course is unbiased. So we should use that. -Now, for the common solution, we have .A distribution of only non-negative values, which should have .expectation of zero .positive, non-zero standard deviation One can can simply see: such distribution is impossible, it cannot exist!
Regards Joerg Bergmann
