For most usual applications, these parameters are automatically generated from the button “Create a command file" of the main menu. However, the command file could also be edited in a subsequent phase (change of parameters values, replacement of some default values, addition of a new step, etc.)
Parameters
naxe: number of requested axes.
nvedi: number of eigenvalues (histogram).
ledin: printing the coordinates of the
individuals (0=no, 1=act, 2=tot).
naxed: number of printed coordinates.
niter: number of iterations (case of iterative
algorithm) (0=no).
nictl: number of "controlled iterations" (0=no).
nbase: dimension of the approximation subspace.
nsimu: number of bootstrap replication
(0 = no bootstrap, max = 30)
nboot: type of bootstrap replication
( 0 = "total bootstrap", 1= partial
bootstrap [default])
ndata: printing the replicated data tables
(1 = yes, 0 = no [default])
Parameters
naxe: number of principal coordinates.
ledit: parameter for printing the rows
(0=no, 1=yes).
ngraf: number of requested graphics
npage: number of pages of the graphics
nrows: number of lines of the graphics
nsimu: number of bootstrap replication
(0 = no bootstrap, max = 30)
nboot: type of bootstrap replication
0 = "total bootstrap",
[bootstrapping the words]
1 = partial bootstrap [default]
[bootstrapping the words]
2 = does not apply here
3 = total specific bootstrap
[bootstrapping the respondents]
4 = partial specific bootstrap
[bootstrapping the respondents]
ndata: printing the replicated data tables
(1 = yes, 0 = no [default])
Parameters
nqexa: number of variables in the dictionary.
nxmod: maximal number of categories
for a variable.
ledit: printing mode of the dictionary
(0=no, 1=short, 2=long)
nidi: number of groups of 4 characters
(identifier of individuals) (0=no).
niexa: number of individuals in file ndonz.
test: value given to missing numerical data (default=.999).
nedit: number of individuals in a check
printout (0=no).
Parameters
ityp: type of textual data file ntexz
(tex=1, enq =2)
nbqt: number of questions per individual
nlig: number of responses to be printed for
each text (0 = only the titles)
lirep: indicates implicit end of a response
(1= 1 line (row)/1 response)
Parameters
naxe: number of requested principal coordinates
ledit: printing the responses
(0 = no, 1 = coor, 2 = tot)
ngraf: number of requested graphics
npage: number of pages of the graphics
nrows: number of lines of the graphics
nbase: dimension of the approximation space
(0= main core diagonalization)
niter: number of iterations (if nbase > 0)
Parameters
naxe: number of described axes
lign: simultaneous listing of the rows
(0=no, 1=yes).
nmax: number of printed "extreme elements"
for each principal axis
Parameters
ledit: printing identifiers of individuals
(0 = no printing, 1 = identifiers of
respondents are printed, default = 0)
lspb: use of a file nspb (0=no, 1=yes)
--- key-word of headings : forme end
Parameters
ledit: printing of words (0=no, 1=nspc, 2=tot).
lclas: printing sorted words (0=no, 1=yes).
--- key-word of headings : delet equiv end
Parameters
ednom: printing the tables (classes * questions)
(0=no).
lpnom: describing partition with questions
(0=no, 1=yes).
cnomi: describing classes with questions (0=no).
pmoda: describing partition with categories (0=no).
cmoda: describing classes with categories (0=no).
pcmin: minimum relative weight (%) for a category.
lsupr: characteristic category if
%(cat./class) > %(cat./total)
(0=no, 1=yes).
pfreq: describing partition with frequencies (0=no).
lpfre: printing thetable (classes * frequencies)
(0=no, 1=act, 2=tot).
cfreq: describing classes with frequencies (0=no).
edcon: describing partition with numerical variables
(0=no, 1=yes).
ccont: describing classes with numerical variables
(0=no).
Parameters
ednom: printing the tables (classes * questions)
(0=no).
lpnom: describing partition with questions
(0=no, 1=yes).
cnomi: describing classes with questions (0=no).
pmoda: describing partition with categories (0=no).
cmoda: describing classes with categories (0=no).
pcmin: minimum relative ( % ) weight for a category.
lsupr: characteristic category if
%(cat./class) > %(cat./total)(0=no,1=yes).
pfreq: describing partition with frequencies (0=no).
lpfre: printing the table (classes * frequencies)
(0=no, 1=act, 2=tot).
cfreq: describing classes with frequencies (0=no).
edcon: describing partition with numerical variables
(0=no, 1=yes).
ccont: describing classes with numerical variables
(0=no).
paxe: describing partition with axes (0=no).
caxe: describing classes with axes(0=no).
ledin: printing correspondence classes*individuals
(0=no,1=compo, 2=affec, 3=tot).
Parameters
seuil: threshold for selecting most significant items.
lcrim: criterion of selection for the categories
( coordinate or test-value)
vtmin: test-value threshold for selecting categories.
Parameters niter: Maximum number of iterations acdif: Maximum difference between required and obtained percentages liste: Printing the individual weights (0=no, 1=yes). ledit: Printing the future provisional dictionnary (0=no , 1=short, 2=long). nxvrd: Maximal number of categorical variables used for the iterative fitting key word: end (after lists of target [or: required] percentages)
Parameters nomot: number of charac. words for each text. norep: number of charac. responses for each text.
Parameters
nvsel: index of the categorical variable defining
the groupings of texts
ledit: parameter for printing the table words*texts
(0=no, 1=yes).
Parameters
naxe: number of requested axes.
pcmin: threshold for "cleaning" of the
active categories (in percent).
lburt: printing the Burt table
(0=no, 1=mass, 2=tot, 3=prof).
nvedi: number of eigenvalues (histogram).
naxed: number of printed coordinates.
ledin: printing the coordinates of individuals
(0=no, 1=act, 2=tot).
ledco: printing the correlations variable-axers
(0=no, 1=yes).
nsimu: number of bootstrap replication
(1 = no bootstrap)
nboot: (1 = partial bootstrap [default])
(0 = total bootstrap. In such a case,
the "cleaning of rare categories"
[parameter pcmin above] is not allowed)
Parameters
coef: technical coefficient
(frequencies of word-lengths)
(default value = 1; put coef=10 for large corpora).
nseu: frequency threshold of the kept words
nxlet: maximum number of letters per word.
nxmax: maximum number of distinct words.
nxsig: maximum number of first letters of word.
ledit: printing the words (0=no, 1=alpha, 2=tri, 3=tot).
nmomi: threshold for the length of words
--- key-word of headings
weak strong accent histo end
Parameters
nxpar: maximum number of requested dendrogram cut.
niter: number of "consolidation" iterations (0=no).
ledin: printing the correspondence classes-individuals
(0=no; 1=class-->individuals;
2=individuals-->classes; 3= both 1 and 2).
naxed: number of axes for printed test-values (0=no).
Parameters
and individuals (rows) )
ledit: printing the number of permuted elements
(0=no, 1=yes).
Parameters ngraf: number of requested graphics npage: number of pages of the graphics nrows: number of lines of the graphics
Parameters
naxe: number of printed axes
ngraf: number of requested graphics
npage: number of pages for the graphics
nrows: number of lines for the graphics
zoom: maximum distance to the origin
(as numbers of standard deviations)
Parameters
lcorr: analysis of the correlation matrix
(0=no, 1=yes).
naxe: number of requested axes
nvedi: number of eigenvalues (histogram).
ledin: printing the coordinates of the individuals
(0=no, 1=act, 2=tot).
naxed: number of printed coordinates.
navir: number of digits (right to the dot)
(printing of the coordinates).
nsimu: number of bootstrap replication
(0 = no bootstrap)
nboot: type of bootstrap replication
(0 = "total bootstrap", 1= partial
bootstrap, 2 = bootstrap on variables)
Parameter
norep: number of characteristic responses
for each text
Parameters
naxu: number of kept axes.
nterm: number of kept terminal elements of tree.
nived: number of levels in histogram of indices
(0=no).
lterm: printing the composition of terminal elements
(0=no, 1=yes).
naxed: number of coordinates for printing terminal
elements (0=no).
ldesc: describing the nodes of the tree
(0=no, 1=yes).
ldend: printing the dendrogram
(0=no, 1=dense, 2=large).
Parameters
lsupr: missing value
( =0 : individuals with m.v. are kept)
( =1 : individuals with m.v. discarded)
( default = 1)
lstat: printing statistics (=1) or not (default=0)
lmat: printing covariance matrix
(default = 0 = no)
(1 = covariance, 2 = correlation)
Then: regression formulae: e.g.:
V8 = V9 + V57 (one per line)
(V8, V9, V57 = continuous variables)
V8 = V67 + V12*V15
(V8 = continuous, V67, V12, V15 = categorical)
Ending the commands with the key word "end"
Parameters nxlon: maximum length of the segment nxseg: maximum number of requested segments nseug: frequency threshold of segments( general) nseu2: frequency threshold of segments of length 2 nseu3: frequency threshold of segments of length 3 nxle2: maximum number of segments of length 2
Parameters
lseli: mode of selection of individuals
(0=all, 1=list, 2=filt).
imass: index of variable 'weight of individuals'
(0=uniform weights).
lzero: coding missing responses (0=norec, 1=rec).
ledit: printing dictionary of selected variables
(0=no, 1=short,2=long).
nxfil: maximum number of constants used in filter
(lseli = filt ou 2).
The selection commands use following key words :
act ill end nomi cont freq
Parameters
numq: index of the selected question
if numq = -1 or numq = list : several questions
will be merged (list of questions numbers
must follow next line)
lseli: mode of selection of the individuals
(0=tot, 1=list, 2=filt).
nxfil: maximum number of constantes used in filter
(lseli = filt or 2).
ldona: presence of files ndona and ndica
(files produced by step ardat)
Parameters
naxu: number of kept axes.
niter: number of iterations for each basic partition.
lmeth: type of clustering (1=mobil,2=stabl).
seuil: number of kept terminal elements
lresi: building of a resisual cluster (1=yes,0=no).
nived: number of levels in the histogram of indices
(0=no).
ldesc: describing nodes of the tree (0=no, 1=yes).
ldend: printing dendrogram (0=no, 1=dense, 2=large).
Parameters
nseu: threshold of frequency for selecting words.
nmomi: minimum number of letters of a kept word.
nremi: minimum number of words of a kept response.
nxmsu: maximum number of words explicitly eliminated.
ledit: printing the dictionaries
(0=no, 1=new, 2=tot).
Parameters
ledit: printing identifiers of supplementary texts.
--- key-word of headings
forme text end
Parameters
lcont: printing elementary statistics about numerical
variables (0=no, 1=yes).
lhist: printing the histograms of numerical variables
(0=no, 1=yes).
unit: default unit for histograms.
zmin: minimal default lower bound for histograms.
zmax: maximal default lower bound for histograms.
nclas: default number of classes for histograms.
lcom: commands for numerical variables (0=no, 1=yes).
nxvdi: number maximum of values for a discretized
numerical variable.
--- key-words of commands
zmin zmax unit nclas hist libre disc end
Parameters
imass: index of the variable 'weight of individuals'
(0=uniform weight generated = default).
lzero: coding missing responses
(0= norec = missing values ignored - default).
(1 = rec = m.v. become a new category)
lfilt: use of a filter (0 = default = no filter)
(1 = a filter will follow)
ledic: printing variables dictionary
(0 = default = no, 1 = short, 2 = large)
letyp: printing standard deviation
(case of a table of mean values)
(0 = no = default)
(1 = yes: means and std dev. are printed)
lnull: printing columns whose weight is zero
(0 = no, 1 = yes = default)
Parameter ledit: printing the table words*texts (0=no, 1=yes).