1993/05/31 by David Seibert · 1 citation
Mathematics · Physics and Astronomy · #Statistical and numerical algorithms #hep-lat #hep-ph #nucl-th
paper · pdf · doi:10.1103/physrevd.49.6240
published as Phys.Rev. D49 (1994) 6240-6243 · CERN-TH.6892/93 (revised), replaced because of major improvements (finding a test for stability of covariance matrix regression and an alternative method for fitting correlated data)
arxiv created 1993/09/16 · openalex publication_date 1994/06/01 · arxiv updated 2009/11/30 · openalex created_date 2016/06/24 · openalex updated_date 2026/07/28
Regression with \ensuremathχ2 constructed from covariance matrices should not be used for some combinations of covariance matrices and fitting functions. Using the technique for unsuitable combinations can amplify systematic errors. This amplification is uncontrolled, and can produce arbitrarily inaccurate results that might not be ruled out by a \ensuremathχ2 test. In addition, this technique can give incorrect (artificially small) errors for fit parameters. I give a test for this instability and a more robust (but computationally more intensive) method for fitting correlated data.