A function minimization computer package (MFIT) for nonlinear parameter estimation providing readily accessible maximum likelihood estimates
✍ Scribed by Jane B. Hazelrig; Malcolm E. Turner Jr.; Eugene Ackerman
- Publisher
- Elsevier Science
- Year
- 1978
- Tongue
- English
- Weight
- 735 KB
- Volume
- 11
- Category
- Article
- ISSN
- 0010-4809
No coin nor oath required. For personal study only.
✦ Synopsis
MFIT is a computer software package for obtaining nonlinear parameter estimates. It was designed, in part, to motivate the maximum likelihood method by making it as accessible to the scientifically oriented user as nonlinear regression or least squares analysis. The approach taken was that of supplying a library of likelihood function subprograms or modules each of which links the user-supplied module to any of four standard function minimization or optimization algorithms available in MFIT. The likelihood module selected by the user evaluates the negative log likelihood given known values for the observations, y, and user-supplied values for the unknown parameters of the likelihood function, q. The user module merely evaluates q = ((p, x) for a known set of experimental conditions, x, and trial values, assigned by the optimization module, for the unknown parameters B.