Abstract: The method of quasi-optimal weights provides a comprehensive, asymptotically optimal, transparent and flexible alternative to the least squares and maximum likelihood methods. The optimality holds for a general non-linear, non-gaussian case.
Author's comment: The outrageous process of so-called anonymous peer review is based on faulty, stupid logic and is simply harmful to Science. The harm inflicted thereby is not negligible at all, and clearly outweighs any benefits to the research community (if such exist). This conclusion is based on my almost 30 years' experience in basic research, and I have never, ever received an anonymous report that would not be superficial and useless at best. Moreover, the anonymous peer review was instrumental in a large-scale plagiaristic affair in the '80-90s that torpedoed a big theory of mine with the most credit gone to the plagiarists by the time my papers were published (and that story is far from being over yet). I am convinced that it is a systemic phenomenon, facilitated by the anonymous peer review system.
If a recruitment committee wish to assess a candidate in such a stupid fashion — why should I be bothered?
Nor should the concerns of commercial journals affect me in any way. What serious researcher needs them with the arXiv anyway?
There is little one can do about these things — but one should try. This is why this paper is not intended to contribute to perpetuation of the ignominious practice of anonymous peer review.
If an expert wishes to communicate their comments to me, welcome — but not through the anonymous peer review channel.
In the Fall of 2006 I learned from a translation of the New Yorker article by Sylvia Nasar and David Gruber about Grigory Perelman's refusing to submit his now famous solution of the Poincare problem to journals. Perelman's reasons seems to be same as mine, essentially: the too low standards of scientific ethics in the international scientific community. Amen.
My journal policies.