alternatives to the least squares solution to peelle’s pertinent puzzle替代最小二乘解peelle的相关难题.pdfVIP

Algorithms 2011, 4, 115-130; doi:10.3390/a4020115 OPEN ACCESS algorithms ISSN 1999-4893 /journal/algorithms Article Alternatives to the Least Squares Solution to Peelle’s Pertinent Puzzle Tom Burr ,⋆, Todd Graves , Nicolas Hengartner , Toshihiko Kawano , Feng Pan and Patrick Talou Statistical Sciences, Los Alamos National Laboratory, Los Alamos, NM 87545, USA; E-Mail: tgraves@ Information Sciences, Los Alamos National Laboratory, Los Alamos, NM 87545, USA; E-Mail: nickh@ Nuclear and Particle Physics, Los Alamos National Laboratory, Los Alamos, NM 87545, USA; E-Mails: kawano@ (T.K.); talou@ (P.T.) Decision Applications, Los Alamos National Laboratory, Los Alamos, NM 87545, USA; E-Mail: fpan@ ⋆ Author to whom correspondence should be addressed; E-Mail: tburr@; Tel.: +1-505-665-7865; Fax: +1-505-667-4470. Received: 28 March 2011; in revised form: 1 June 2011 / Accepted: 8 June 2011 / Published: 23 June 2011 Abstract: Peelle’s Pertinent Puzzle (PPP) was described in 1987 in the context of estimating fundamental parameters that arise in nuclear interaction experiments. In PPP, generalized least squares (GLS) parameter estimates fell outside the range of the data, which has raised concerns that GLS is somehow flawed and has led to suggested alternatives to GLS estimators. However, there have been no corresponding performance comparisons among methods, and one suggested approach involving simulated data realizations is statistically incomplete. Here we provide performance comparisons among estimators, introduce