+--------------------------------------------------------------------+
 |                                                                    |
 |                        MONITORING OF SEARCH                        |
 |                                                                    |
 +--------------------------------------------------------------------+

 MEANING: History report on the Estimation Step search
 CONTEXT: NONMEM output

 DISCUSSION:
 NONMEM  output  includes  a report of the history of the search under-
 taken in the Estimation Step for parameter estimates.  This report  is
 called  the  intermediate  output from the Estimation Step, because it
 consists of summaries of the progress of the search, from iteration to
 iteration, and because it may be viewed as the search progresses, pro-
 vided the NONMEM output file may be viewed as the  search  progresses.
 This report can also be viewed in a special (unbuffered) file.

 The  search  is  for  parameter  values that minimize the value of the
 OBJECTIVE FUNCTION  (See minimum value of objective function).   Basi-
 cally, it entails the following steps.

 1.   The  search  is  carried out in a different parameter space.  The
      parameters are transformed to unconstrained parameters (UCP).  In
      the  transformation  process a scaling occurs so that the initial
      estimate of each of the UCP is 0.1.  (NONMEM Users Guide  I  Sec-
      tions  C.3.5.1 "ESTIMATION" and G.3 "Messages from the Estimation
      Step" refer to the UCP as "scaled transformed parameters" or STP.
      NONMEM  Users Guide II Section F "Rescaling" refers to the UCP as
      "rescaled canonical parameters" or RCP.)

 2.   At the  current parameter estimate the GRADIENT vector (i.e., the
      vector  of  first  partial  derivatives of the objective function
      with respect to the UCP) is  computed.   An  approximate  Hessian
      matrix  (See hessian)  is  also  computed.  An ITERATION SUMMARY,
      including the current parameter estimate and the gradient vector,
      may placed into the intermediate output.

      With  NONMEM  7,  the  parameter  estimates are also displayed in
      their natural (unscaled) space. These  lines  are  identified  as
      NPARAMETR  and precede the PARAMETER lines, which display the UCP
      values.  Note that when OMEGA (or SIGMA)  have  block  structure,
      the  values  in  NPARAMETR  are listed in upper triangular order,
      whereas elsewhere in NONMEM output they are  displayed  in  lower
      triangular  order.   For  example, suppose the NONMEM output dis-
      plays a 3x3 OMEGA matrix as

      OMEGA(1,1)
      OMEGA(2,1) OMEGA(2,2)
      OMEGA(3,1) OMEGA(3,2) OMEGA(3,3)

      It will be listed in NPARAMTR as:

      OMEGA(1,1) OMEGA(2,1) OMEGA(3,1) OMEGA(2,2) OMEGA(3,2) OMEGA(3,3)

      This is inconsistent, but it cannot be changed; option  ORDER  of
      the $ESTIMATION record does not affect the NPARAMTR line.

 3.   Using  the  gradient  vector  and  Hessian matrix, a direction in
      parameter space, emananting from the current parameter  estimate,
      is  computed, and a search is undertaken along this direction for
      an approximate minimum point.  When this point is  found,  NONMEM
      returns  to step 2.  (An ITERATION consists of the computation of
      the direction, the search along the direction, and  the  computa-
      tion of the gradient vector and Hessian matrix at the approximate
      minimum point.)

 4.   Iteration stops when any one of the following conditions holds:

      A successful (local) minimum point has been found.

      The maximum number of FUNCTION EVALUATIONS allowed  by  the  user
      (MAXEVALS option of $ESTIM) is exceeded.

      It was not possible to successfully locate a minimum point due to
      so-called ROUNDING ERRORS.

 Here is an example of intermediate output from the Estimation Step:

 MONITORING OF SEARCH:
  ITERATION NO.:    0    OBJECTIVE VALUE:   110.244034784025  NO. OF FUNC. EVALS.:   6
  CUMULATIVE NO. OF FUNC. EVALS.:        6
  NPARAMETR:  3.0000E+00  8.0000E-02  4.0000E-02  6.0000E+00  5.0000E-03
  3.0000E-01  2.0000E-04  6.0000E-03  4.0000E-01  4.0008E-01
  PARAMETER:  1.0000E-01  1.0000E-01  1.0000E-01  1.0000E-01  1.0000E-01
  1.0000E-01  1.0000E-01  1.0000E-01  1.0000E-01  1.0000E-01
  GRADIENT:   1.4640E+01 -1.6691E+01  5.6460E+01 -9.4129E+00 -1.1699E+01
  1.0623E+01 -6.3485E-01 -4.7803E+01 -9.6457E+00  5.2414E+00

  ITERATION NO.:    5    OBJECTIVE VALUE:   105.453694506861  NO. OF FUNC. EVALS.:   7
 CUMULATIVE NO. OF FUNC. EVALS.:       45
  NPARAMETR:  2.8166E+00  8.0433E-02  3.8563E-02  6.6123E+00  9.7862E-03
 -3.9509E-02  2.3438E-04  8.1375E-03  4.9657E-01  4.0355E-01
  PARAMETER: -5.3055E-02  1.0703E-01  5.7598E-02  1.4858E-01  1.8644E-01
 -1.2545E-02  1.5794E-01  1.3452E-01  3.7714E-02  1.0431E-01
  GRADIENT:   3.5103E-01 -1.3214E+01  2.8003E+01  1.1115E+00  1.9193E+00
 -5.3252E+00  7.1533E-02 -2.2091E+01 -1.4424E+00  8.2329E+00

  ITERATION NO.:   10    OBJECTIVE VALUE:   104.747695563098  NO. OF FUNC. EVALS.:   7
  CUMULATIVE NO. OF FUNC. EVALS.:       80
  NPARAMETR:  2.7362E+00  7.6935E-02  3.5982E-02  5.7362E+00  1.0840E-03
 -2.9208E-01  2.3701E-04  9.4013E-03  5.1805E-01  3.8858E-01
  PARAMETER: -1.1928E-01  4.9234E-02 -2.3022E-02  7.7517E-02  2.2174E-02
 -9.9574E-02  1.9498E-01  1.4956E-01 -1.7173E-01  8.5414E-02
  GRADIENT:  -1.3933E+00 -7.4632E+00  9.3144E+00  1.0280E+00  9.6395E-01
 -3.1313E+00  1.0212E+00 -5.3784E+00 -2.0828E+00 -9.1477E-01

  ITERATION NO.:   15    OBJECTIVE VALUE:   104.561086502423  NO. OF FUNC. EVALS.:   7
  CUMULATIVE NO. OF FUNC. EVALS.:      116
  NPARAMETR:  2.7735E+00  7.8097E-02  3.6292E-02  5.5466E+00  5.1735E-03
 -1.2911E-01  2.3941E-04  9.0947E-03  5.1461E-01  3.8760E-01
  PARAMETER: -8.8567E-02  6.8686E-02 -1.2994E-02  6.0711E-02  1.0762E-01
 -4.4762E-02  1.9028E-01  1.4643E-01 -8.4016E-02  8.4157E-02
  GRADIENT:   7.7937E-03  5.7189E-03  3.0426E-02 -5.1648E-03 -1.2939E-02
  1.4490E-02  7.4602E-04 -2.5227E-02 -3.1443E-03  1.5799E-02

 ITERATION NO.:   19    OBJECTIVE VALUE:   104.561067398632  NO. OF FUNC. EVALS.:   0
  CUMULATIVE NO. OF FUNC. EVALS.:      149
  NPARAMETR:  2.7739E+00  7.8129E-02  3.6307E-02  5.5498E+00  5.2393E-03
 -1.2767E-01  2.3985E-04  9.1064E-03  5.1521E-01  3.8742E-01
  PARAMETER: -8.8304E-02  6.9219E-02 -1.2510E-02  6.0998E-02  1.0895E-01
 -4.4249E-02  1.9095E-01  1.4652E-01 -8.3269E-02  8.4127E-02
  GRADIENT:   1.2144E-04 -1.4883E-04 -6.6257E-06 -7.5745E-05 -1.6157E-05
  1.1307E-04  6.4128E-05  1.1193E-04 -1.7655E-05  4.2676E-05

 Note that the values of the PARAMETERs are the values of the  UCP,  so
 that at the 0th iteration, all the PARAMETERs have the value 0.1.

 The  first  parameter  (and gradient) elements correspond to the THETA
 elements which are not fixed.  The remaining  elements  correspond  to
 the  OMEGA and SIGMA elements which are not fixed, but not in a simple
 1-1 manner unless OMEGA and SIGMA are constrained to be diagonal.

 The PRINT option of the $ESTIMATION record determines how often itera-
 tion  summaries  are printed: not at all (with PRINT=0);  only for the
 0th and last iterations (with PRINT=9999); for the 0th iteration,  for
 every  10th  iteration  thereafter,  and  for the last iteration (with
 PRINT=10, as illustrated above).

 If a model specification file is output, then the estimates  may  also
 be  seen  in  the original parameterization for those iterations whose
 summaries appear in intermediate output.  These estimates may be found
 in  file INTER in the same order and format as elsewhere in the NONMEM
 output.  With NONMEM 7, INTER exists after the run is finished.        |

 When the Estimation Step terminates, it reports its success or lack of
 it, as in this example:

 MINIMIZATION SUCCESSFUL
 NO. OF FUNCTION EVALUATIONS USED:      149
 NO. OF SIG. DIGITS IN FINAL EST.:  4.7

 Each  (UCP)  element of the mimimum point is determined to a number of
 significant digits.  The number of significant digits reported is  the
 number  of  significant  digits  in the least-well-determined element.
 The report "MINIMIZATION SUCCESSFUL" is issued when this number is  no
 less than the number of significant digits requested using the SIGDIG-
 ITS option of the $ESTIMATION record.  Note  that  this  report  alone
 does not assure that a global (or even a local) minimum point has been
 located; what appears to be a minimum point may  be  a  saddle  point.
 Nor, if a minimum point has been located, does the report alone assure
 that the objective function is not "flat" in a region  of  the  point.
 For such assurances, one also needs to implement the Covariance Step.

REFERENCES: Guide I Section C.3.5.1
REFERENCES: Guide V Section 10.4.1


  
Go to main index.
  
Created by nmhelp2html v. 1.0 written by Niclas Jonsson (Modified by AJB 5/2006,11/2007,10/2012)