Multivariate least squares fitting program using modified Gram-Schmidt transformations

✍ Scribed by M.J. O'Connell

• Publisher: Elsevier Science
• Year: 1974
• Tongue: English
• Weight: 870 KB
• Volume: 8
• Category: Article
• ISSN: 0010-4655
• DOI: 10.1016/0010-4655(74)90085-x

✦ Synopsis

Nature of physical problem issue)

Fitting a dependent variable in terms of functions of many in-Computer: Installation: dependent variables, e.g. finding the momentum of a charged IBM 370/195 Rutherford Laboratory particle from observation of its trajectory in a magnetic field. ICL 1906A Atlas Laboratory Method of solution CDC 6600 CERN Residual sum of squares is minimized using modified Gram-Operating system: IBM OS/HASP Schmidt orthogonal basis vectors. Programming language used: FORTRAN Restrictions on the complexity of the problem None in principle; coding arranged for 6 independent variables, High speed store required: 37.5 K words 250 observations, up to 1000 trial functions. No. of bits in a word: 32 Typical running time Overlay structure.' None 15 seconds on IBM 360/195 No. of magnetic tapes required: 1 for input Unusual features of the program User has considerable control over the functions to be includ-Other peripherals used: Card reader, lineprinter ed, with the facility to reject marginally contributing terms. No. of cards in combined program and test deck: 1411