NON LINEAR LEAST SQUARES COMPUTER ANALYSIS OF MOSSBAUER TRANSMISSION SPECTRA,
Abstract
It is known that computer programs which utilize blind search and maximum slope methods to fit a non-linear N-parameter function to experimental Mossbauer data are slow and unreliable, and if N is 'large,' say N > 10 they cannot be used. The Hartley-modified Gauss-Newton method for fitting nonlinear functions by least squares and was applied to the fitting of Mossbauer data. Using this modified Gauss-Newton method, computer running time was reduced by roughly a factor of 20, and it was possible to fit functions with as many as 30 parameters in a few minutes.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1967
- Accession Number
- AD0654929
Entities
People
- Bruce L. Chrisman
- Thomas A. Tumolillo
Organizations
- University of Illinois Urbana–Champaign