N.I.M.R.O.D.  
Functions/Subroutines

likelihood.f90 File Reference

Go to the source code of this file.

Functions/Subroutines

subroutine vraisobs (b, num_pat1, m)
 Compute Individual observed log-likelihood.
subroutine vraistot (NDIM2, X, NF2, FUNVLS)
 Compute Individual observed log-likelihood conditionnally to random effects.
subroutine vraistotEXP (NDIM2, X, NF2, FUNVLS)
 Compute Individual observed likelihood conditionnally to random effects.
subroutine ODEschedule
 Schedule the ODE solver time points.

Function Documentation

subroutine ODEschedule ( )

Schedule the ODE solver time points.

AUTHOR : Melanie Prague Daniel Commenges Julia Drylewicz Jeremy guedj Rodolphe Thiebaut

DESCRIPTION :

Find observation rim eto run the ODE solver at these points

CAUTION : Normally, these subroutines must not be modified in routine by users.

MODIFICATION:

01/09/2012 - Prague - Refactoring

INFORMATIONS:

Definition at line 340 of file likelihood.f90.

References WorkingSharedValues::listtime, and WorkingSharedValues::tailletime.

Referenced by EBCurves(), and nimrod().

Here is the caller graph for this function:

subroutine vraisobs ( double precision,dimension(m),intent(in)  b,
integer,intent(in)  num_pat1,
integer,intent(in)  m 
)

Compute Individual observed log-likelihood.

AUTHOR : Melanie Prague Daniel Commenges Julia Drylewicz Jeremy guedj Rodolphe Thiebaut

DESCRIPTION :

The individual likelihood ( $\mathcal{L}^{\theta}_i$) is computed by integrating over the random effects via the adaptive Gaussian quadrature .

MODIFICATION :

01/09/2012 - Prague - Refactoring

INFORMATIONS:

Parameters:
[in]bparameter vector
[in]mparameter vector lenth
[in]num_pat1Identifier/Key of the patient considered

Definition at line 32 of file likelihood.f90.

References WorkingSharedValues::abserrfuncpa, WorkingSharedValues::adaptive2, WorkingSharedValues::b1, WorkingSharedValues::likelihoodERROR, WorkingSharedValues::likelihoodPRECISION, WorkingSharedValues::numpat1, WorkingSharedValues::vrais_obs, and vraistot().

Referenced by funcpa().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine vraistot ( integer,intent(in)  ndim2,
double precision,dimension(ndim2),intent(in)  X,
integer,intent(in)  nf2,
double precision,intent(out)  funvls 
)

Compute Individual observed log-likelihood conditionnally to random effects.

AUTHOR : Melanie Prague Daniel Commenges Julia Drylewicz Jeremy guedj Rodolphe Thiebaut

DESCRIPTION :

For each patient, the individual log-likelihood given the random effects ( $L^{\theta}_{\mathcal{F}_i|u_i}$) is computed. It is a function of $g_m(X^i(t_{ij}))$ computed by the ODE solver since the $Y^i_{m}(t_{ij})$ are independent Gaussian variables.

MODIFICATION :

01/09/2012 - Prague - Refactoring

INFORMATIONS:

Parameters:
[in]NDIM2Number of random effects.
[in]NF2Number of components of the integral for the adaptive gaussian quadrature. Generally=1.
[in]Xestimated individual random effects
[out]funvlsfunction value

Definition at line 104 of file likelihood.f90.

References WorkingSharedValues::adaptive2, WorkingSharedValues::b1, WorkingSharedValues::censor, WorkingSharedValues::detersauv, WorkingSharedValues::numpat1, mpimod::numproc, Constante::pigrec, WorkingSharedValues::scaleinv2sauv, WorkingSharedValues::scaleinvsauv, solution(), WorkingSharedValues::startsauv, and WorkingSharedValues::systeme.

Referenced by vraisobs().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine vraistotEXP ( integer,intent(in)  ndim2,
double precision,dimension(ndim2),intent(in)  X,
integer,intent(in)  nf2,
double precision,intent(out)  funvls 
)

Compute Individual observed likelihood conditionnally to random effects.

In some cases, computation of the Likelihood is easier than computation of the loglikelihood because of the computer digits precision.

AUTHOR : Melanie Prague Daniel Commenges Julia Drylewicz Jeremy guedj Rodolphe Thiebaut

DESCRIPTION :

For each patient, the individual likelihood given the random effects ( $\mathcal{L}^{\theta}_{\mathcal{F}_i|u_i}$) is computed. It is a function of $g_m(X^i(t_{ij}))$ computed by the ODE solver since the $Y^i_{m}(t_{ij})$ are independent Gaussian variables.

MODIFICATION :

01/01/2013 - Prague - Creation of the function

INFORMATIONS:

Parameters:
[in]NDIM2Number of random effects.
[in]NF2Number of components of the integral for the adaptive gaussian quadrature. Generally=1.
[in]Xestimated individual random effects
[out]funvlsfunction value

Definition at line 222 of file likelihood.f90.

References WorkingSharedValues::adaptive2, WorkingSharedValues::b1, WorkingSharedValues::censor, WorkingSharedValues::detersauv, WorkingSharedValues::numpat1, mpimod::numproc, Constante::pigrec, WorkingSharedValues::scaleinv2sauv, WorkingSharedValues::scaleinvsauv, solution(), WorkingSharedValues::startsauv, and WorkingSharedValues::systeme.

Referenced by derivVRAISTOT(), and funcpaRandomEffect().

Here is the call graph for this function:

Here is the caller graph for this function: