The examples described in the book by
Some additional examples are available to illustrate
some particular uses of nls2.
The model files and the data-creation
commands files are
in the directory data of
the package.
The execution-commands are in the directory demo.
Names of the files are: <example-name>.R
See also the examples list sorted by function-names.
The examples are:
Name | in chapters | illustrated functions | characteristics |
---|---|---|---|
beetles | 6.7: Binomial non-linear models | nls2 | introduced in the 2nd edition
numerical equalities constraints |
cheese | 7.1.8: Multinomial non-linear models | nls2, confidence. | introduced in the 2nd edition |
corti | 1.6: Nonlinear regression model and parameter estimation | nls2 | numerical equalities constraints |
2.6: Accuracy of estimators, confidence intervals and tests | confidence | ||
4.6: Diagnostics of model misspecification | nls2, plres, plvar | ||
5.6: Calibration and Prediction | calib | ||
7.2.4: Poisson non-linear models | nls2 | ||
ej | nls2 |
This example works when the Nag library is available, only.
The loadnls2 command should be customised
according to the localisation of the
Nag library on the site.
The programs are not automatically generated but written "at hand" and use Nag subroutines. This is an example of estimation by three models with common parameters. Each model is calculated on a particular subset of data: two are calculated by C-programs which can be automatically generated, the third one is calculated by a Fortran program calling Nag subroutines. Values are transmitted to these programs via Fortran COMMON. | |
elisa | 1.6: Nonlinear regression model and parameter estimation | pldnls2, nls2 | with curves,
equality constraints between parameters and 'binf' or 'bsup' bounds |
2.6: Accuracy of estimators, confidence intervals and tests | wald, nls2, confidence, conflike, plfit, bootstrap | ||
fbeetles | 6.7: Binomial non-linear models | nls2, confidence. | introduced in the 2nd edition
with curves, numerical equality constraints between parameters |
insect | 6.7: Binomial non-linear models | nls2, confidence. | introduced in the 2nd edition
'binf' or 'bsup' bounds on parameters |
isomer | 1.6: Nonlinear regression model and parameter estimation | nls2, plfit | 'binf' or 'bsup' bounds on parameters |
2.6: Accuracy of estimators, confidence intervals and tests | confidence, ellips, iso, nls2, bootstrap | ||
4.6: Diagnostics of model misspecification | plfit, plres. | ||
mag | nls2, renls2. |
bootstrap estimation on the Cortisol example.
Works when the Nag library is available, only.
The model is written in C and calls Fortran subroutines from the Nag library to generate pseudo-observations. | |
miners | 7.1.8: Multinomial non-linear models | nls2, confidence. | introduced in the 2nd edition |
nasturnium | 5.6: Calibration and Prediction | nls2, bootstrap, calib | |
orobranche | 6.7: Binomial non-linear models | nls2, bootstrap. | introduced in the 2nd edition |
ovo | 1.6: Nonlinear regression model and parameter estimation | nls2, plfit | with curves, model described by a odes, numerical equality constraints between parameters |
2.6: Accuracy of estimators, confidence intervals and tests | ellips, iso. | ||
4.6: Diagnostics of model misspecification | plres, plvar. | ||
ownmethod | nls2. | numerical equality constraints between parameters
and 'binf' or 'bsup' bounds.
Example of an estimation method defined by the user. You can use this method in case of errors auto-correlation. This file contains two examples: example of simulated data and example of real data. (Reference). | |
pasture | 1.6: Nonlinear regression model and parameter estimation | pldnls2, nls2, plfit | |
2.6: Accuracy of estimators, confidence intervals and tests | confidence, nls2, bootstrap | ||
4.6: Diagnostics of model misspecification | nls2, plfit. | ||
5.6: Calibration and Prediction | confidence, bootstrap. | ||
pept | 3.6: Variance estimation | nls2, pldnls2, plres, plfit, conflike, confidence. | alternated method of estimation |
4.6: Diagnostics of model misspecification | nls2, plres, plfit. | ||
perm | nls2, calcmodnls2. | with curves and weights,
model described by a odes | |
pg | nls2, pldnls2, plres, plit, plfit. | with curves, numerical equality constraints between parameters | |
root | 4.6: Diagnostics of model misspecification | pldnls2, nls2, plres, plfit. | |
sim | 4.6: Diagnostics of model misspecification | nls2, confidence, plfit. | |
simul | nls2, calcmodnls2. | model calculated by a Fortran program. The regression function is solution of an implicit equation. Works when the Nag library is available, only. | |
substrat | nls2, calcmodnls2. | with curves, model described by a odes numerical and equality constraints between parameters and 'binf' or 'bsup' bounds | |
tiller | 3.6: Variance estimation | nls2, confidence, conflike, bootstrap. | changed in the 2nd edition |
vaso | 6.7: Binomial non-linear models | nls2. | introduced in the 2nd edition
equality constraints between parameters explicit 'family': Bernoulli |
volterra | nls2, calcmodnls2. | model described by a odes | |
weibull | nls2, rexnls2. |
Reference for the example ownmethod:
D. Deschamps,
Rapport de Stage DESS Mathématiques Appliquées,
INRA, Jouy-en-Josas, juin 1994.
Creation: 2003 -