5 %PussIP%  
Iteration %s% $ @ IMPIP0 2 @
6 %PussIP:PROMPT%  Stopping criterion: %critere% $ @ IMPIP1 2 @
198 %PussIP:PROMPT%  Optimal step: %omega% $ @ IMPIP12 2 @
196 %PussIP:PROMPT%  Sigma**2: %sigma% $ @ IMPIP14 2 @
7 %PussIP:PROMPT%  Statistical criterion: %critere% $ @ IMPIP2 2 @
8 %PussIP:PROMPT%  Value of the Gauss-Marquardt parameter (lambda): %lambda% $ @ IMPIP3 2 @
10 %PussIP:PROMPT%  Fitted values of the regression function: $ @ IMPIP4 1 @
197 %PussIP:PROMPT%  Value of the direction: $ @ IMPIP13 1 @
154 %PussIP:PROMPT%  Fitted values of FOdes: 
  (for each observation, for each value of time, n values, where n is the number of equations).
 $ @ IMPIP10 1 @
155 %PussIP:PROMPT%  Fitted values of the derivatives of FOdes: $ @ IMPIP11 1 @
11 %PussIP:PROMPT%  Fitted values of the variance: $ @ IMPIP5 1 @
12 %PussIP:PROMPT%  Values of %B-ou-Eta-ou-D%: $ @ IMPIP6 2 @
13 %PussIP:PROMPT%  Values of the derivatives of the regression function with respect to the parameters: 
 $ @ IMPIP7 1 @
14 %PussIP:PROMPT%  Values of the derivatives of the variance with respect to the %regression-ou-variance% parameters: 
 $ @ IMPIP8 2 @
98 %PussIp.PROMPT%  Current values of the %regression-ou-variance% parameters:
  $ @ IMPIP9 2 @
15   %ErrIp% : %s%-st error when evaluating the %s% model:
  code %s%, step %s%, iteration %s%.
 $ @ IMPERRIP1 6 @
3 %ErrIp% : The error occurred when calculating %s% at observation: $ @ IMPERRIP2 2 @
4 %ErrIp% :  Current values of the active %regression-ou-variance% parameters: $ @ IMPERRIP3 2 @
20 %CModErr% : %n%-st attempt at evaluating the model.
 $ @ WARCM4 2 @
149 %PussIter% : %n%-st attempt succeeded.
 $ @ IMPIT1 2 @
188  %nlfonc.h% : The result cannot be obtained with arguments: %s% %s%.
 The NaN (NotANumber) value is returned. $ @ WARNUM 3 @
1 %NLEtape% 
 ++++++++++
 + Step %i% +
 ++++++++++
 $ @ IMPDEBUT 2 @
2 %NLEtape% 
 +++++++++++++++++
 + End of step %i% +
 +++++++++++++++++
 $  @ IMPFIN 2 @
9 %PussIter%
   ------------------------------- 
   End of the iterative process
   because convergence is reached.
   ------------------------------- 
$ @ IMPIT0 1 @
99 %PussIter%
   ------------------------------- 
   No iterative process
   because convergence is reached on the initial values
   ----------------------------------------------------- 
$ @ IMPIT00 1 @
17 %PussIter%
  ----------------------------------------------------
  End of the iterative process
  because the maximal number of iterations is reached.
  ----------------------------------------------------
$  @ IMPIT2 1 @
65 %PussIter%
  -----------------------------------------------------
  End of the iterative process
  because the maximal number of times for which the 
  minimisation criterion has been reached consecutively 
  at the beginning of an interval is reached.
  -----------------------------------------------------
$  @ IMPIT3 1 @
74 %PussIter% 
  --------------------------------------------
  End of the iterative process
  because the model could not be calculated.
  --------------------------------------------
$ @ IMPIT4 1 @
18 %CModErr% : The %s% model could not be calculated in step %s%.
  The iterative process is stopped.
  New values for the parameters have been found, but the model
  could not be calculated for them. You can try again with inequality
  numerical constraints or with another parametrisation.
$  @ WARCM2 3 @
156 %CModErr% : The regression function could not be calculated
  at the starting values.
  The iterative process is stopped.
  You can try again with other starting values, another set of data
  or another model. Look at the arguments of divisions, log, etc...
$ @ ERRMODF 1 @
157 %CModErr% : The variance function could not be calculated
  at the starting values.
  The iterative process is stopped.
 You can try again with other starting values or another set of data.
$ @ ERRMODV 1 @
19 %CModErr% : The maximal number of errors ("max.err.c1"= %s%) is reached.
  The iterative process is stopped; the output structure contains the last valid values.
$  @ WARCM3 2 @
22 %PussIter% : The maximal number of iterations ("max.iters" = %s%) is reached.
  The iterative process is stopped.
$ Be careful: the results are meaningful only if the stopping criterion
  is small. If it is not the case, try again with the last estimated
  values as starting values and with a greater number of iterations. @ WARPUSS2 2 @
23   %PussIter% : The maximal number of times ("max.err.c2" = %s%) for which the 
  minimisation criterion has been reached consecutively at the beginning of an 
  interval is reached.
  The iterative process is stopped.
  You can try again with the Gauss-Marquardt algorithm, or with a greater value
  of the initial value of the parameter of this algorithm, or with other 
  starting values or with another model. 
$ @ WARPUSS3 2 @
61   %PussIter% : The minimisation criterions are equal at the 3 points of calculation,
  even after %s% ("max.err.c2") attempts.
  The iterative process is stopped.
 You can try again with the Gauss-Marquardt algorithm, or with a greater value
  of the bound of the parameter of the GM algorithm. 
$ @ WARPUSS4 2 @
21 %PussIter% 
  The model cannot be calculated: the maximum number of times the direction is 
  modified when the model cannot be calculated ("max.err.c1" = %s%) is reached.
$ @ WARPUSS1 1 @
147  %CCov2.PROMPT% INTERMEDIATE RESULTS
  ---------------------------------------
  Elements from which is calculated the
  asymptotic variance in step %s%.
  ---------------------------------------
$ @ IMPCOV 2 @
90 %NLVMod% : Value affected to variance type ("vari.type") : CST (constant).
$ When the variance is weighted, there must be weights in the data.
 @ WARVMOD1 1 @
24 %NLVCtxPuss% : Value affected to algorithm type ("algorithm"): %s%.
  (%s% = GM and %s% = GN).
$ This is the default value or the provided value was not valid.
 @ WARVPUSS1,4 @
25 %NLVCtxPuss% : Value affected to the term by which direction
  is multiplied when an error occurred during the model evaluation ("omega.c1"): %s%.
$ This value must be positive and less than 1.
 @ WARVPUSS2,2 @
26 %NLVCtxPuss% : Initial value affected to parameter lambda ("lambda.start"): %s%.
$ The initial value of the parameter of the Gauss-Marquardt algorithm
  must be positive.
 @ WARVPUSS3,2 @
27 %NLVCtxPuss% : Value affected to the term by which parameter lambda
  is multiplied when calculating the optimal step if the criterion is 
  minimum at the center of the current interval ("lambda.c1"): %s%.
$ This value must be positive and less than 1.
 @ WARVPUSS4,2 @
28 %NLVCtxPuss% : Value affected to the term by which parameter lambda
  is multiplied when calculating the optimal step if the criterion is 
  minimum at the beginning of the current interval ("lambda.c2"): %s%.
$ This value must be greater than 1 divided by the term by which lambda
  is multiplied at each iteration.
 @ WARVPUSS5,2 @
32 %NLVCtxPuss% : Maximal value affected to parameter lambda ("max.lambda"): %s%.
$ This value must be positive and less than the initial value of lambda.
 @ WARVPUSS9,2 @
29 %NLVCtxPuss% : Maximal value affected to the stopping criterion ("max.stop.crit"): %s%.
$ This value must be positive or null.
 @ WARVPUSS6,2 @
30 %NLVCtxPuss% : Maximal value affected to the number of times the direction 
  is modified when the model cannot be calculated ("max.err.c1"): %s%.
$ This value must be positive or null.
 @ WARVPUSS7,2 @
161 %NLVCtxPuss% : Maximal value affected to the number of attempts the model
  is evaluated at the beginning of an interval ("max.err.c2"): %s%.
$ This value must be greater than 1.
 @ WARVPUSS12,2 @
31 %NLVCtxPuss% : Maximal value affected to the number of iterations ("max.iters"): %s%.
$ This is the default value (calculated depending on the number of parameters)
  because there is no "max.iters" argument or the provided value is too great or negative.
  @ WARVPUSS8,2 @
33 %NLVCtxPuss% : Value affected to the correction of the optimal step if the criterion 
  is minimum at the beginning point of the interval and if GN algorithm is used ("omega.c2"): %s%.
$ This value must be positive.
 @ WARVPUSS10,2 @
34 %NLVCtxPuss% : Value affected to the type of sigma ("sigma2.type"): %s%.
  (%s% = KNOWN, %s% = VARREP, %s% =VARRESID, %s% = IGNORED, %s% = VARINTRA).
$ This is the default value because there is no "sigma2.type" argument or
  the provided value is not valid.
 @ WARVPUSS11,7 @
35 %VerifMu% : Value affected to the type of moments ("mu.type"): %s%.
  (%s% = KNOWN, %s% = MUGAUSS, %s% = MURES, %s% = MURESREP).
$ This is the default value because there is no "mu.type" argument or
   the provided value is not valid.
 @ WARMU1,6 @
36 %VerifMu% : The type of moments ("mu.type") was MURESREP but the minimum number 
  of replications (%s%) is not enough (less than %s%).
  So, the type of moments is set to %s%.
  (%s% = KNOWN, %s% = MUGAUSS, %s% = MURES, %s% = MURESREP).
$ @ WARMU2,4 @
39 %NLVCtxPuss% : The type of sigma ("sigma2.type") was VARREP but, in that case,
  the number of replications at each observation must be at least %s%.
  So, the type of sigma is set to MURES.
$ @ WARVPUSS16,2 @
40 %NLVCtxPuss% : The type of sigma ("sigma2.type") is VARREP but be careful:
  there is one or several observations for which the number of replications
  is less than the desirable minimum number (%s%).
$ @ WARVPUSS17,2 @
41 %NLVCtxNum% : The estimation method for step %e% is valid only when there is 
  one step (i.e, simultaneous estimation of regression and variance parameters).
  This step - and the next ones, if any - are ignored. 
$ @ WARES1 2 @
162 %NLVCtxPuss% : The family method %e% is wrong. 
$ @ WARVPUSS18 2 @
163 %NLVCtxPuss% : The name of the variable "n" is missing (nameN) in the context.
$ @ WARVPUSS19 1 @
42 %NLVCtxNum% : The estimation method for even step %e% is a method for
  estimating regression parameters only:
  This step - and the next ones, if any - are ignored.
$ @ WARES2 2 @
44 %NLVCtxNum% : The estimation method for odd step %e% is a method for
  estimating the parameters that appear in variance only.
  This step - and the next ones, if any - are ignored.
$ @ WARES4 2 @
43 %NLVCtxNum% : The estimation method is a method for estimating regression parameters
  but variance (model[["vari.type"]]) depends on other parameters.
  Execution is stopped after completion of all the checks.
  You can try again by setting equality numerical constraints on variance parameters. 
$ @ WARES3 1 @
46 %NLVCtxNum% : The estimation method for step %e% is a method for estimating the
  parameters that appear in variance only but there are no such parameters:
  This step - and the next ones, if any - are ignored.
$ @ WARES6 2 @
47 %NLVCtxNum% : The context for step %e% is not valid:
  This step - and the next ones, if any - are ignored.
$ @ WARES7 2 @
48 %NLVCtxNum% : The estimation method for step %e% is not valid:
  This step - and the next ones, if any - are ignored.
$ @ WARES8 2 @
49 %EstDefaut% : Value affected to the estimation method for step %e%: %s%.
  (1=MLTB, 2=MLSTB/QLTB, 4=MLT, 5=WLST, 6=OLST, 7=MLST/QLT, 8=VITWLS, 9=OLSB, 10=MLSB/QLB).
$ This is the default value.
 @ WARESD1 3 @
50 %EstTheta% : The estimation method for step %e% was WLST but the variance (model[["vari.type"]])
  depends on regression parameters.
  So, the estimation method is set to OLST.
$ @ WAREST1 2 @
51 %EstTheta% : The estimation method for step %e% was MLST or QLT but the variance (model[["vari.type"]])
  does not depend on regression parameters.
  So, the estimation method is set to WLST if the variance is weighted and to OLST if not.
$ @ WAREST2 2 @
60 %EstTheta% : The estimation method for step %e% was VITWLS but the variance should not be 
  estimated with replications (model[["vari.type"]] is not VI):
  So, the estimation method is set to WLST if the variance (model[["vari.type"]]) is weighted 
  and to OLST if not.
$ @ WAREST5 2 @
52 %EstTheta% : The estimation method for step %e% was VITWLS but, in that case,
  the number of replications at each observation must be at least %s%.
  So, the estimation method is set to OLST.
$ @ WAREST3 3 @
53 %EstTheta% : The estimation method for step %e% is VITWLS but be careful:
  there is one or several observations for which the number of replications
  is less than the desirable minimum number (%s%).
$ @ WAREST4 3 @
54 %EstSimul% : The estimation method for step %e% was a method for estimating
  simultaneously regression parameters and parameters that appear in variance only 
  but the variance (model[["vari.type"]]) is constant.
  So, the estimation method is set to OLST.
$ @ WARESS1 2 @
55 %EstSimul% : The estimation method for step %e% was a method for estimating
  simultaneously regression parameters and parameters that appear in variance only
  but the variance (model[["vari.type"]]) is weighted.
  So, the estimation method is set to WLST.
$ @ WARESS2 2 @
56 %EstSimul% : The estimation method for step %e% was a method for estimating
  simultaneously regression parameters and parameters that appear in variance only
  but there are no parameters that appear in variance only.
  So, the estimation method is set to MLT if it was MLTB and to MLST (resp. QLT) if it was MLSTB (resp. QLTB). 
$ @ WARESS3 2 @
255 %VerifMien% : For step %e% (your own method) the way of calculating the statistical criterion (%s) is wrong.
  (1=LOGV, 2=STOPCRIT, 3=NWSST, 4=VWSS, 5=IVWSS, 6=NWSSB, 7=SIGMA2, 11=MYOWN).
$ @ WARVFM1 3 @
256 %VerifMien% : The estimation method for even step %e% is your own method
  but the way of calculating the statistical criterion (%s) is possible only 
  for an odd step (estimation of parameters that appear in variance only).
$ @ WARVFM2 3 @
257 %VerifMien% : The estimation method for odd step %e% (estimation of parameters
  that appear in variance only) is your own method, but the way of calculating the
  statistical criterion (%s) is possible only for an even step.
$ @ WARVFM3 3 @
258 %VerifMien% : The estimation method for step %e% is your own method and the 
  way of calculating the statistical criterion is %s, but be careful:
  there is one or several observations for which the number of replications
  is less than the desirable minimum number (%s%).
$ @ WARVFM4 4 @
259 %VerifMien% : The estimation method for step %e% is your own method and the
  way of calculating the statistical criterion is %s, but the variance type ("vari.type")
  is not a intra-replications variance.
$ @ WARVFM5 2 @
310 %NLVCtxInteg% : Invalid 'print' option in the integration context.
  Its value is set to %s%.
$ @ WAINDLO  2 @
311 %NLVCtxInteg% : Invalid 'itol' option in the integration context.
  Its value is set to %s%.
$ @ WARITOL  2 @
312 %NLVCtxInteg% : Invalid 'atol' option in the integration context.
  Its value is set to %s%.
$ @ WARATOL  2 @
313 %NLVCtxInteg% : Invalid 'rtol' option in the integration context.
  Its value is set to %s%.
$ @ WARRTOL  2 @
314 %NLVCtxInteg% : Invalid 'jacobian.meth' option in the integration context.
  Its value is set to %s%.
$ @ WARJT  2 @
315 %NLVCtxInteg% : Invalid 'iopt' option in the integration context.
  Its value is set to %s%.
$ @ WARIOPT  2 @
127 %CCorr% : The asymptotic variance ("as.var") could not be calculated and
  therefore the correlation matrix ("correlation").
$ @ WARCORR 1 @
128 %CVarZmv% : The %s%-st diagonal term of the variance matrix of Z is negative 
  or null (%s%). 
  So, all the outputs depending on Z are not calculated, i.e:
  "B.varZ.B", "as.var" and "correlation".
 It is probably due to a wrong estimation of the moments of order 3 and 4.
  You can try again with setting the type of the moments ("mu.type") to MURESREP if 
  there are replications and to MUGAUSS if not.
$ @ WARVARZ1 3 @
101 %CCov2% : This program is not able to calculate the variance in step 2
  when Eta is not defined as in the maximum likelihood case.
$ @ WARCCOV2 1 @
102 %CCov3% : To calculate the variance in step 3, Eta must be 
  obtained in step 2 by the function corresponding to the maximum 
  likelihood case and in step 3 by the function corresponding to the
  maximum likelihood or ordinary least squares case.
$ @ WARCCOV3 1 @
37 %VEgalP% : The %s%-st equality constraint between the parameters of %s%
  is equal to: %s%.
  It must be positive and less than the total number of parameters, i.e: %s%.
  Execution is stopped after completion of all the checks.
$ @ WARVEGALP1 5 @
38 %VEgalP% : The equality constraints between the parameters of %s% does not
  include parameter %s%.
  Execution is stopped after completion of all the checks.
$ @ WARVEGALP2 3 @
57 %EstMien% : The estimation method in step %e% is your own method
  but the dimension of the exhaustive statistics (%d%) is not positive.
  Execution is stopped after completion of all the checks.
$ @ WARESM1 3@
58 %EstMien% : The estimation method in step %e% is your own method
  but the form of the matrix to be inversed (%f%) is not valid.
  Execution is stopped after completion of all the checks.
$ @ WARESM2 6 @
59 %EstMien% : The estimation method in step %e% is your own method
  but its efficacity (%f%) is not valid.
  Execution is stopped after completion of all the checks.
$ @ WARESM3 5 @
91 %NLVMod% : When the variance (model[["vari.type"]]) is constant, 
  estimated by the weighted residual sum of squares
  or the intra-replications variance, do not define a variance function.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD2 1 @
45 %NLVCtxNum% : The context is entirely wrong.
  Execution is stopped after completion of all the checks.
$ @ WARES5 1 @
77 %PreparVar% : The dimension, %s%, of the independent variables is wrong.
  Execution is stopped after completion of all the checks.
$ @ WARDON1 2 @
78 %PreparVar% : The dimension, %s%, of the weighting vector is wrong.
  Execution is stopped after completion of all the checks.
$ @ WARDON2 2 @
75 %PreparVar% : The dimension, %s%, of the vector of curves is wrong.
  Execution is stopped after completion of all the checks.
$ @ WARDON3 2 @
89 %PreparVar% : The dimension, %s%, of the vector that contains the names of the 
   observations is wrong.
  Execution is stopped after completion of all the checks.
$ @ WARDON4 2 @
96 %PreparVar% : The dimension, %s%, of the the vector that contains the names of 
   the independent variables is wrong.
  Execution is stopped after completion of all the checks.
$ @ WARDON5 2 @
79 %PreparVar% : The %s%-st value of the vector of weights is negative.
  Execution is stopped after completion of all the checks.
$ @ WARPOIDS1 2 @
80 %NLVMod% : When the variance is neither constant, nor estimated by the weighted 
  residual sum of squares or the intra-replications variance (model[["vari.type"]]), 
  the variance function must be defined.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD3 1 @
81 %NLVMod% : There are parameters that appear in variance only; these parameters
  must be defined.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD4 1 @
82 %NLVMod% : The dimension of the starting values vector for %s% 
  parameters (%s%) must be equal to the total number of parameters of the model of %s%
  multiplied by the number of curves, i.e %s%.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD5 5 @
83 %NLVMod% : The dimension of the vector that contains the second level parameters
  of %s% (%s%) must be equal to the number of these parameters in the formal description
  of the model, i.e: %s%.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD6 5 @
84 %NLVMod% : The dimension of the constraint vector %s% on the parameters 
  of %s% (%s%) must be equal to the total number of parameters of %s%, i.e: %s%.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD7 6 @
86 %NLVMod% : The %s%-st lower bound constraint on the parameters
  of %s% (%s%) is greater than the corresponding starting value (%s%).
  Execution is stopped after completion of all the checks.
$ @ WARVMOD9 5 @
87 %NLVMod% : The %s%-st upper bound constraint on the parameters
  of %s% (%s%) is less than the corresponding starting value (%s%).
  Execution is stopped after completion of all the checks.
$ @ WARVMOD10 5 @
97 %NLVMod% : The %s%-st upper bound constraint on the parameters
  of %s% (%s%) is equal to the lower bound (%s%).
  Execution is stopped after completion of all the checks.
$ @ WARVMOD15 5 @
88 %NLVMod% : The %s%-st numerical equality constraint on the parameters
  of %s% (%s%) is not equal to the corresponding starting value (%s%).
  Execution is stopped after completion of all the checks.
$ @ WARVMOD11 5 @
92 %NLVMod% : There is a constraint of type %s% on the %s%-st parameter of 
  %s% (%s%) which is not equal with the constraint set on the %s%-st parameter 
  (which is equal to the %s%-st), i.e: %s%.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD12 8 @
93 %NLVMod% : You declare parameters that appear in the variance only in the model
  description; so, a variance function must be defined.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD13 1 @
94 %NLVMod% : The starting value of the %s%-st parameter of %s% is %s%,
  but the starting value of the %s%-st parameter (which is equal to the %s%-st) 
  is different, i.e: %s%.
  Execution is stopped after completion of all the checks.
$ @ WARVMOD14 7 @
95 %NLVDonM% : The number of observations for the %s%-st curve (%s%) is less
  than the number of basic parameters, i.e: %s%.
  Execution is stopped after completion of all the checks.
$ @ WARVDONM 4 @
110 %NLVCrole% : Invalid component %s% in the 'control' argument (%s%).
  Valid values:  TRUE (1) or FALSE (0).
  Execution is stopped after completion of all the checks.
$ @ WARVCO1 3 @
111 %NLVCrole% : Invalid component %s% in the 'control' argument (%s%).
  It must not be negative.
  Execution is stopped after completion of all the checks.
$ @ WARVCO2 3 @
62 %ChangerDim% : All the weights are null.
  Execution is stopped after completion of all the checks.
$ @ WARPOIDS2 1 @
159 %NLVCtxPuss% : The number of requested steps (%s%) must be positive and 
  less than %s%.
  Execution is stopped after completion of all the checks.
$ @ ERRVPUSS 2 @
121 %VerifContBeta% : When the variance (model[["vari.type"]]) depends on regression 
  parameters and on other parameters and when the estimation method implies 
  estimation of the regression parameters only, you must fix all the variance 
  parameters or proceed by steps.
  Execution is stopped after completion of all the checks.
$ @ WARCONTB 1 @
112 %NL% : Invalid trace length ("trace.length") in the 'control' argument (%s%).
  It is greater than the maximum (%s%) or negative.
$ @ ERRTRACE 3 @
120 %NLVInit% : End of execution:
      ----------------
  Invalid inputs.
$ @ ERRVINIT 1 @
148 %PussIter% : Invalid statistical criterion type (%s%).
  Execution is stopped.
$ @ ERRTYPES 2 @
150 %matherr% : An error occurred when calculating a mathematical function:
  Type of error: %s% , name of the function: %s%,
  first argument: %s%, possibly, second argument: %s%, result: %s%.
$ @ ERRMATH 6 @
105 %CMu% : Invalid type of moments ("mu.type" = %s%).
  Valid values are: %s% (KNOWN), %s% (MUGAUSS), %s% (MURES), %s% (MURESREP).
  Execution is stopped.
$ @ ERRCMU 2 @
66 %VerifMu% : The dimension of the vectors containing the moments must be equal to the
  total number of data, i.e %s%.
  Execution is stopped.
$ @ ERRMU 2 @
73 %CDirGM% : The linear system cannot be solved.
  Execution is stopped.
 You can try again with the Gauss-Marquardt algorithm, or with a greater value
  of the initial value of the parameter of this algorithm, or with other
  starting values or with another model.
$ @ ERRSYS 1 @
67 %ActAMod% : An "overflow" occurred when transforming the %s%-st active parameter
  into the %s%-st 'effective' parameter.
  (the 'effective' parameters are obtained by removing the equal parameters).
  (value of the parameter: %s%, lower bound: %s%).
  Execution may be stopped.
$ @ ERROVFL1 5 @
68 %ActAMod% : An "overflow" occurred when transforming the %s%-st active parameter
  into the %s%-st 'effective' parameter.
  (the 'effective' parameters are obtained by removing the equal parameters).
  (value of the parameter: %s%, upper bound: %s%).
  Execution may be stopped.
$ @ ERROVFL2 5 @
69 %ActAMod% : An "overflow" occurred when transforming the %s%-st active parameter
  into the %s%-st 'effective' parameter.
  (the 'effective' parameters are obtained by removing the equal parameters).
  (value of the parameter: %s%, lower bound: %s%, upper bound: %s%).
  Execution may be stopped.
$ @ ERROVFL3 6 @
70 %ModAAct% : An "overflow" occurred when transforming the %s%-st 'effective' parameter
  into the %s%-st active parameter.
  (the 'effective' parameters are obtained by removing the equal parameters).
  (value of the parameter: %s%, lower bound: %s%, upper bound: %s%).
  Execution may be stopped.
$ @ ERROVFL4 6 @
16 %DModAAct% : An "overflow" occurred when transforming into the 'effective' dimension
  the derivative with respect to the %s%-st active parameter.
  (the 'effective' parameters are obtained by removing the equal parameters).
  (value of the derivative: %s%, lower bound of the parameter: %s%, 
   upper bound: %s%).
  Execution may be stopped.
$ @ ERROVFL5 5 @
71 %CDirec% : The %s%-st term of the diagonal of W is negative or null (%s%).
  Execution may be stopped.
 This error occurs when calculating the direction or the stopping criterion:
  before inversion, each term of W, W(a,b), is normalized by the squared root
  of the diagonal term: W(a,a)*W(b,b).
$ @ ERRDIREC 3 @
72 %MatInv% : We are trying to inverse a singular matrix.
  Execution may be stopped.
 This error occurs when calculating the direction or the stopping criterion:
  Matrix W is singular.
  You can try again with the Gauss-Marquardt algorithm, or with a greater value
  of the initial value of the parameter of this algorithm, or with other starting 
  values or with another model.
$  @ ERRCALC 1 @
145 %CVariance% : The %s%-st value calculated by the variance function is negative ou null.
  Execution may be stopped.
$ @ ERRVAR 2 @
158 %CVariance% : The %s%-st value calculated by the variance function is negative or null
  at the first iteration.
  Execution may be stopped.
$ @ ERRVAR1 2 @
661 %nlmacros% : Error when allocating memory space for the structure called %nom%.
  Execution is stopped.
$ Probably, there is not enough working memory space.
@ ERRALLOC 2 @
301 %NLVCtxInteg% The number of parameters of the differential equations system is negative. $ @ ERNBTHET 1 @
302 %NLVCtxInteg% If the number of parameters of the differential equations system is null, 
  the initial conditions must be parameters to be estimated. $ @ ERTYPCI 1 @
303  %NLVCtxInteg% Error on 'IndicX' or 'IndicCi' in the structure returned by the
 'analder' system: probably an error in the formal description of the model.
   $ @ ERINDLO 1 @
304 %NLVCtxInteg% There are not enough or too many equations (according to BSUPFICT):
  probably an error in the formal description of the model. $ @ ERNBEQUA 1 @
305 %NLVCtxInteg% Error on 'IndicTj' in the structure returned by the
 'analder' system: probably an error in the formal description of the model.
 $ @ ERINITJ 1 @
306 %NLVCtxInteg% Error in the dimension of the initial conditions. $ @ ERLCI 1 @
307 %CalcInteg% Error detected by LSODA: code %s% (see the LSODA manual). $ @ ERLSODA 2 @
199 %Init3MCPT% The variance function cannot be calculated at the last
 estimated values of the parameters. $ @ ERRV3 1 @
400 %NLVTrace% Error of the trace length (GNLControle.LgTrace). $ @ WALGTRACE 1 @


