#include <Paramest.hh>
Inheritance diagram for RazorBack::Paramest_:

Public Types | |
| enum | Weighting_ { EQUAL, RELATIVE, USER } |
| Weighting_: equal, relative, user-defined. More... | |
Public Methods | |
| Paramest_ () | |
| Init to empty. | |
| Paramest_ (size_t NM, size_t NP) | |
| Init to desired sizes if known in advance Actual regression calls may change the settings. | |
| void | weight (Weighting_ W) |
| set the weighting scheme. | |
| const Vecdbl_& | param () const |
| Returns the estimated parameters. | |
| const Vecdbl_& | sdev () const |
| Returns the estimated standard deviations of the parameters. | |
| const Smatdbl_& | correl () const |
| Returns the estimated correlation matrix between the parameters. | |
| float | tcrit_95 () const |
| tcrit_95(DOF) returns the value of the t-distribution at 95% significance level with Nf degrees of freedom. More... | |
Protected Methods | |
| void | resize () |
| double | normalise () |
| double | solve (Vecdbl_ &Sol, double Lambda0, double Lambdafact) |
| void | sdev_correl (double Q) |
Protected Attributes | |
| size_t | Nm |
| size_t | Np |
| size_t | Nf |
| no. of measurements, no.of params, degrees of freedom. | |
| Smatdbl_ | Xtx |
| Fisher's observation matrix. | |
| Vecdbl_ | Xty |
| rhs of linreg eqn. | |
| Vecdbl_ | Norm |
| normalisation vector. | |
| Weighting_ | Weighting |
| weight type. | |
| Vecdbl_ | P |
| Vecdbl_ | Sdev |
| parameter values and standard deviations. | |
| Smatdbl_ | Correl |
| parameter correlation matrix. | |
Static Protected Methods | |
| bool | posdef_inv (Smatdbl_ &A) |
Definition at line 39 of file Paramest.hh.
|
|
Weighting_: equal, relative, user-defined.
Definition at line 45 of file Paramest.hh. |
|
|
Init to empty.
Definition at line 62 of file Paramest.hh. |
|
|
Init to desired sizes if known in advance Actual regression calls may change the settings.
Definition at line 69 of file Paramest.hh. |
|
|
set the weighting scheme.
Definition at line 74 of file Paramest.hh. |
|
|
Returns the estimated parameters.
Definition at line 77 of file Paramest.hh. |
|
|
Returns the estimated standard deviations of the parameters.
Definition at line 80 of file Paramest.hh. |
|
|
Returns the estimated correlation matrix between the parameters.
Definition at line 83 of file Paramest.hh. |
|
|
tcrit_95(DOF) returns the value of the t-distribution at 95% significance level with Nf degrees of freedom. Use for confidence intervals. |
|
|
Definition at line 95 of file Paramest.hh. Referenced by Paramest_(). |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Definition at line 50 of file Paramest.hh. |
|
|
Definition at line 50 of file Paramest.hh. |
|
|
no. of measurements, no.of params, degrees of freedom.
Definition at line 50 of file Paramest.hh. |
|
|
Fisher's observation matrix.
Definition at line 51 of file Paramest.hh. |
|
|
rhs of linreg eqn.
Definition at line 52 of file Paramest.hh. |
|
|
normalisation vector.
Definition at line 53 of file Paramest.hh. |
|
|
weight type.
Definition at line 54 of file Paramest.hh. |
|
|
Definition at line 55 of file Paramest.hh. |
|
|
parameter values and standard deviations.
Definition at line 55 of file Paramest.hh. |
|
|
parameter correlation matrix.
Definition at line 56 of file Paramest.hh. |
1.2.6 written by Dimitri van Heesch,
© 1997-2001